diff --git a/samples/timm/inception_v4/graph_hash.txt b/samples/timm/inception_v4/graph_hash.txt
new file mode 100644
index 000000000..9d895b1fe
--- /dev/null
+++ b/samples/timm/inception_v4/graph_hash.txt
@@ -0,0 +1 @@
+bedc93b3e52234d31236e0a996facacac3f14888bd77f9cdd36e55fe76b2b061
\ No newline at end of file
diff --git a/samples/timm/inception_v4/graph_net.json b/samples/timm/inception_v4/graph_net.json
new file mode 100644
index 000000000..1373fe3b5
--- /dev/null
+++ b/samples/timm/inception_v4/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/inception_v4/input_meta.py b/samples/timm/inception_v4/input_meta.py
new file mode 100644
index 000000000..e69de29bb
diff --git a/samples/timm/inception_v4/input_tensor_constraints.py b/samples/timm/inception_v4/input_tensor_constraints.py
new file mode 100644
index 000000000..e69de29bb
diff --git a/samples/timm/inception_v4/model.py b/samples/timm/inception_v4/model.py
new file mode 100644
index 000000000..e65984cca
--- /dev/null
+++ b/samples/timm/inception_v4/model.py
@@ -0,0 +1,5215 @@
+import torch
+
+
+class GraphModule(torch.nn.Module):
+    def forward(
+        self,
+        L_self_modules_features_modules_0_modules_conv_parameters_weight_: torch.nn.parameter.Parameter,
+        s1: torch.SymInt,
+        L_x_: torch.Tensor,
+        L_self_modules_features_modules_0_modules_bn_buffers_running_mean_: torch.Tensor,
+        L_self_modules_features_modules_0_modules_bn_buffers_running_var_: torch.Tensor,
+        L_self_modules_features_modules_0_modules_bn_parameters_weight_: torch.nn.parameter.Parameter,
+        L_self_modules_features_modules_0_modules_bn_parameters_bias_: torch.nn.parameter.Parameter,
+        L_self_modules_features_modules_1_modules_conv_parameters_weight_: torch.nn.parameter.Parameter,
+        L_self_modules_features_modules_1_modules_bn_buffers_running_mean_: torch.Tensor,
+        L_self_modules_features_modules_1_modules_bn_buffers_running_var_: torch.Tensor,
+        L_self_modules_features_modules_1_modules_bn_parameters_weight_: torch.nn.parameter.Parameter,
+        L_self_modules_features_modules_1_modules_bn_parameters_bias_: torch.nn.parameter.Parameter,
+        L_self_modules_features_modules_2_modules_conv_parameters_weight_: torch.nn.parameter.Parameter,
+        L_self_modules_features_modules_2_modules_bn_buffers_running_mean_: torch.Tensor,
+        L_self_modules_features_modules_2_modules_bn_buffers_running_var_: torch.Tensor,
+        L_self_modules_features_modules_2_modules_bn_parameters_weight_: torch.nn.parameter.Parameter,
+        L_self_modules_features_modules_2_modules_bn_parameters_bias_: torch.nn.parameter.Parameter,
+        L_self_modules_features_modules_3_modules_conv_modules_conv_parameters_weight_: torch.nn.parameter.Parameter,
+        L_self_modules_features_modules_3_modules_conv_modules_bn_buffers_running_mean_: torch.Tensor,
+        L_self_modules_features_modules_3_modules_conv_modules_bn_buffers_running_var_: torch.Tensor,
+        L_self_modules_features_modules_3_modules_conv_modules_bn_parameters_weight_: torch.nn.parameter.Parameter,
+        L_self_modules_features_modules_3_modules_conv_modules_bn_parameters_bias_: torch.nn.parameter.Parameter,
+        L_self_modules_features_modules_4_modules_branch0_modules_0_modules_conv_parameters_weight_: torch.nn.parameter.Parameter,
+        L_self_modules_features_modules_4_modules_branch0_modules_0_modules_bn_buffers_running_mean_: torch.Tensor,
+        L_self_modules_features_modules_4_modules_branch0_modules_0_modules_bn_buffers_running_var_: torch.Tensor,
+        L_self_modules_features_modules_4_modules_branch0_modules_0_modules_bn_parameters_weight_: torch.nn.parameter.Parameter,
+        L_self_modules_features_modules_4_modules_branch0_modules_0_modules_bn_parameters_bias_: torch.nn.parameter.Parameter,
+        L_self_modules_features_modules_4_modules_branch0_modules_1_modules_conv_parameters_weight_: torch.nn.parameter.Parameter,
+        L_self_modules_features_modules_4_modules_branch0_modules_1_modules_bn_buffers_running_mean_: torch.Tensor,
+        L_self_modules_features_modules_4_modules_branch0_modules_1_modules_bn_buffers_running_var_: torch.Tensor,
+        L_self_modules_features_modules_4_modules_branch0_modules_1_modules_bn_parameters_weight_: torch.nn.parameter.Parameter,
+        L_self_modules_features_modules_4_modules_branch0_modules_1_modules_bn_parameters_bias_: torch.nn.parameter.Parameter,
+        L_self_modules_features_modules_4_modules_branch1_modules_0_modules_conv_parameters_weight_: torch.nn.parameter.Parameter,
+        L_self_modules_features_modules_4_modules_branch1_modules_0_modules_bn_buffers_running_mean_: torch.Tensor,
+        L_self_modules_features_modules_4_modules_branch1_modules_0_modules_bn_buffers_running_var_: torch.Tensor,
+        L_self_modules_features_modules_4_modules_branch1_modules_0_modules_bn_parameters_weight_: torch.nn.parameter.Parameter,
+        L_self_modules_features_modules_4_modules_branch1_modules_0_modules_bn_parameters_bias_: torch.nn.parameter.Parameter,
+        L_self_modules_features_modules_4_modules_branch1_modules_1_modules_conv_parameters_weight_: torch.nn.parameter.Parameter,
+        L_self_modules_features_modules_4_modules_branch1_modules_1_modules_bn_buffers_running_mean_: torch.Tensor,
+        L_self_modules_features_modules_4_modules_branch1_modules_1_modules_bn_buffers_running_var_: torch.Tensor,
+        L_self_modules_features_modules_4_modules_branch1_modules_1_modules_bn_parameters_weight_: torch.nn.parameter.Parameter,
+        L_self_modules_features_modules_4_modules_branch1_modules_1_modules_bn_parameters_bias_: torch.nn.parameter.Parameter,
+        L_self_modules_features_modules_4_modules_branch1_modules_2_modules_conv_parameters_weight_: torch.nn.parameter.Parameter,
+        L_self_modules_features_modules_4_modules_branch1_modules_2_modules_bn_buffers_running_mean_: torch.Tensor,
+        L_self_modules_features_modules_4_modules_branch1_modules_2_modules_bn_buffers_running_var_: torch.Tensor,
+        L_self_modules_features_modules_4_modules_branch1_modules_2_modules_bn_parameters_weight_: torch.nn.parameter.Parameter,
+        L_self_modules_features_modules_4_modules_branch1_modules_2_modules_bn_parameters_bias_: torch.nn.parameter.Parameter,
+        L_self_modules_features_modules_4_modules_branch1_modules_3_modules_conv_parameters_weight_: torch.nn.parameter.Parameter,
+        L_self_modules_features_modules_4_modules_branch1_modules_3_modules_bn_buffers_running_mean_: torch.Tensor,
+        L_self_modules_features_modules_4_modules_branch1_modules_3_modules_bn_buffers_running_var_: torch.Tensor,
+        L_self_modules_features_modules_4_modules_branch1_modules_3_modules_bn_parameters_weight_: torch.nn.parameter.Parameter,
+        L_self_modules_features_modules_4_modules_branch1_modules_3_modules_bn_parameters_bias_: torch.nn.parameter.Parameter,
+        L_self_modules_features_modules_5_modules_conv_modules_conv_parameters_weight_: torch.nn.parameter.Parameter,
+        L_self_modules_features_modules_5_modules_conv_modules_bn_buffers_running_mean_: torch.Tensor,
+        L_self_modules_features_modules_5_modules_conv_modules_bn_buffers_running_var_: torch.Tensor,
+        L_self_modules_features_modules_5_modules_conv_modules_bn_parameters_weight_: torch.nn.parameter.Parameter,
+        L_self_modules_features_modules_5_modules_conv_modules_bn_parameters_bias_: torch.nn.parameter.Parameter,
+        L_self_modules_features_modules_6_modules_branch0_modules_conv_parameters_weight_: torch.nn.parameter.Parameter,
+        L_self_modules_features_modules_6_modules_branch0_modules_bn_buffers_running_mean_: torch.Tensor,
+        L_self_modules_features_modules_6_modules_branch0_modules_bn_buffers_running_var_: torch.Tensor,
+        L_self_modules_features_modules_6_modules_branch0_modules_bn_parameters_weight_: torch.nn.parameter.Parameter,
+        L_self_modules_features_modules_6_modules_branch0_modules_bn_parameters_bias_: torch.nn.parameter.Parameter,
+        L_self_modules_features_modules_6_modules_branch1_modules_0_modules_conv_parameters_weight_: torch.nn.parameter.Parameter,
+        L_self_modules_features_modules_6_modules_branch1_modules_0_modules_bn_buffers_running_mean_: torch.Tensor,
+        L_self_modules_features_modules_6_modules_branch1_modules_0_modules_bn_buffers_running_var_: torch.Tensor,
+        L_self_modules_features_modules_6_modules_branch1_modules_0_modules_bn_parameters_weight_: torch.nn.parameter.Parameter,
+        L_self_modules_features_modules_6_modules_branch1_modules_0_modules_bn_parameters_bias_: torch.nn.parameter.Parameter,
+        L_self_modules_features_modules_6_modules_branch1_modules_1_modules_conv_parameters_weight_: torch.nn.parameter.Parameter,
+        L_self_modules_features_modules_6_modules_branch1_modules_1_modules_bn_buffers_running_mean_: torch.Tensor,
+        L_self_modules_features_modules_6_modules_branch1_modules_1_modules_bn_buffers_running_var_: torch.Tensor,
+        L_self_modules_features_modules_6_modules_branch1_modules_1_modules_bn_parameters_weight_: torch.nn.parameter.Parameter,
+        L_self_modules_features_modules_6_modules_branch1_modules_1_modules_bn_parameters_bias_: torch.nn.parameter.Parameter,
+        L_self_modules_features_modules_6_modules_branch2_modules_0_modules_conv_parameters_weight_: torch.nn.parameter.Parameter,
+        L_self_modules_features_modules_6_modules_branch2_modules_0_modules_bn_buffers_running_mean_: torch.Tensor,
+        L_self_modules_features_modules_6_modules_branch2_modules_0_modules_bn_buffers_running_var_: torch.Tensor,
+        L_self_modules_features_modules_6_modules_branch2_modules_0_modules_bn_parameters_weight_: torch.nn.parameter.Parameter,
+        L_self_modules_features_modules_6_modules_branch2_modules_0_modules_bn_parameters_bias_: torch.nn.parameter.Parameter,
+        L_self_modules_features_modules_6_modules_branch2_modules_1_modules_conv_parameters_weight_: torch.nn.parameter.Parameter,
+        L_self_modules_features_modules_6_modules_branch2_modules_1_modules_bn_buffers_running_mean_: torch.Tensor,
+        L_self_modules_features_modules_6_modules_branch2_modules_1_modules_bn_buffers_running_var_: torch.Tensor,
+        L_self_modules_features_modules_6_modules_branch2_modules_1_modules_bn_parameters_weight_: torch.nn.parameter.Parameter,
+        L_self_modules_features_modules_6_modules_branch2_modules_1_modules_bn_parameters_bias_: torch.nn.parameter.Parameter,
+        L_self_modules_features_modules_6_modules_branch2_modules_2_modules_conv_parameters_weight_: torch.nn.parameter.Parameter,
+        L_self_modules_features_modules_6_modules_branch2_modules_2_modules_bn_buffers_running_mean_: torch.Tensor,
+        L_self_modules_features_modules_6_modules_branch2_modules_2_modules_bn_buffers_running_var_: torch.Tensor,
+        L_self_modules_features_modules_6_modules_branch2_modules_2_modules_bn_parameters_weight_: torch.nn.parameter.Parameter,
+        L_self_modules_features_modules_6_modules_branch2_modules_2_modules_bn_parameters_bias_: torch.nn.parameter.Parameter,
+        L_self_modules_features_modules_6_modules_branch3_modules_1_modules_conv_parameters_weight_: torch.nn.parameter.Parameter,
+        L_self_modules_features_modules_6_modules_branch3_modules_1_modules_bn_buffers_running_mean_: torch.Tensor,
+        L_self_modules_features_modules_6_modules_branch3_modules_1_modules_bn_buffers_running_var_: torch.Tensor,
+        L_self_modules_features_modules_6_modules_branch3_modules_1_modules_bn_parameters_weight_: torch.nn.parameter.Parameter,
+        L_self_modules_features_modules_6_modules_branch3_modules_1_modules_bn_parameters_bias_: torch.nn.parameter.Parameter,
+        L_self_modules_features_modules_7_modules_branch0_modules_conv_parameters_weight_: torch.nn.parameter.Parameter,
+        L_self_modules_features_modules_7_modules_branch0_modules_bn_buffers_running_mean_: torch.Tensor,
+        L_self_modules_features_modules_7_modules_branch0_modules_bn_buffers_running_var_: torch.Tensor,
+        L_self_modules_features_modules_7_modules_branch0_modules_bn_parameters_weight_: torch.nn.parameter.Parameter,
+        L_self_modules_features_modules_7_modules_branch0_modules_bn_parameters_bias_: torch.nn.parameter.Parameter,
+        L_self_modules_features_modules_7_modules_branch1_modules_0_modules_conv_parameters_weight_: torch.nn.parameter.Parameter,
+        L_self_modules_features_modules_7_modules_branch1_modules_0_modules_bn_buffers_running_mean_: torch.Tensor,
+        L_self_modules_features_modules_7_modules_branch1_modules_0_modules_bn_buffers_running_var_: torch.Tensor,
+        L_self_modules_features_modules_7_modules_branch1_modules_0_modules_bn_parameters_weight_: torch.nn.parameter.Parameter,
+        L_self_modules_features_modules_7_modules_branch1_modules_0_modules_bn_parameters_bias_: torch.nn.parameter.Parameter,
+        L_self_modules_features_modules_7_modules_branch1_modules_1_modules_conv_parameters_weight_: torch.nn.parameter.Parameter,
+        L_self_modules_features_modules_7_modules_branch1_modules_1_modules_bn_buffers_running_mean_: torch.Tensor,
+        L_self_modules_features_modules_7_modules_branch1_modules_1_modules_bn_buffers_running_var_: torch.Tensor,
+        L_self_modules_features_modules_7_modules_branch1_modules_1_modules_bn_parameters_weight_: torch.nn.parameter.Parameter,
+        L_self_modules_features_modules_7_modules_branch1_modules_1_modules_bn_parameters_bias_: torch.nn.parameter.Parameter,
+        L_self_modules_features_modules_7_modules_branch2_modules_0_modules_conv_parameters_weight_: torch.nn.parameter.Parameter,
+        L_self_modules_features_modules_7_modules_branch2_modules_0_modules_bn_buffers_running_mean_: torch.Tensor,
+        L_self_modules_features_modules_7_modules_branch2_modules_0_modules_bn_buffers_running_var_: torch.Tensor,
+        L_self_modules_features_modules_7_modules_branch2_modules_0_modules_bn_parameters_weight_: torch.nn.parameter.Parameter,
+        L_self_modules_features_modules_7_modules_branch2_modules_0_modules_bn_parameters_bias_: torch.nn.parameter.Parameter,
+        L_self_modules_features_modules_7_modules_branch2_modules_1_modules_conv_parameters_weight_: torch.nn.parameter.Parameter,
+        L_self_modules_features_modules_7_modules_branch2_modules_1_modules_bn_buffers_running_mean_: torch.Tensor,
+        L_self_modules_features_modules_7_modules_branch2_modules_1_modules_bn_buffers_running_var_: torch.Tensor,
+        L_self_modules_features_modules_7_modules_branch2_modules_1_modules_bn_parameters_weight_: torch.nn.parameter.Parameter,
+        L_self_modules_features_modules_7_modules_branch2_modules_1_modules_bn_parameters_bias_: torch.nn.parameter.Parameter,
+        L_self_modules_features_modules_7_modules_branch2_modules_2_modules_conv_parameters_weight_: torch.nn.parameter.Parameter,
+        L_self_modules_features_modules_7_modules_branch2_modules_2_modules_bn_buffers_running_mean_: torch.Tensor,
+        L_self_modules_features_modules_7_modules_branch2_modules_2_modules_bn_buffers_running_var_: torch.Tensor,
+        L_self_modules_features_modules_7_modules_branch2_modules_2_modules_bn_parameters_weight_: torch.nn.parameter.Parameter,
+        L_self_modules_features_modules_7_modules_branch2_modules_2_modules_bn_parameters_bias_: torch.nn.parameter.Parameter,
+        L_self_modules_features_modules_7_modules_branch3_modules_1_modules_conv_parameters_weight_: torch.nn.parameter.Parameter,
+        L_self_modules_features_modules_7_modules_branch3_modules_1_modules_bn_buffers_running_mean_: torch.Tensor,
+        L_self_modules_features_modules_7_modules_branch3_modules_1_modules_bn_buffers_running_var_: torch.Tensor,
+        L_self_modules_features_modules_7_modules_branch3_modules_1_modules_bn_parameters_weight_: torch.nn.parameter.Parameter,
+        L_self_modules_features_modules_7_modules_branch3_modules_1_modules_bn_parameters_bias_: torch.nn.parameter.Parameter,
+        L_self_modules_features_modules_8_modules_branch0_modules_conv_parameters_weight_: torch.nn.parameter.Parameter,
+        L_self_modules_features_modules_8_modules_branch0_modules_bn_buffers_running_mean_: torch.Tensor,
+        L_self_modules_features_modules_8_modules_branch0_modules_bn_buffers_running_var_: torch.Tensor,
+        L_self_modules_features_modules_8_modules_branch0_modules_bn_parameters_weight_: torch.nn.parameter.Parameter,
+        L_self_modules_features_modules_8_modules_branch0_modules_bn_parameters_bias_: torch.nn.parameter.Parameter,
+        L_self_modules_features_modules_8_modules_branch1_modules_0_modules_conv_parameters_weight_: torch.nn.parameter.Parameter,
+        L_self_modules_features_modules_8_modules_branch1_modules_0_modules_bn_buffers_running_mean_: torch.Tensor,
+        L_self_modules_features_modules_8_modules_branch1_modules_0_modules_bn_buffers_running_var_: torch.Tensor,
+        L_self_modules_features_modules_8_modules_branch1_modules_0_modules_bn_parameters_weight_: torch.nn.parameter.Parameter,
+        L_self_modules_features_modules_8_modules_branch1_modules_0_modules_bn_parameters_bias_: torch.nn.parameter.Parameter,
+        L_self_modules_features_modules_8_modules_branch1_modules_1_modules_conv_parameters_weight_: torch.nn.parameter.Parameter,
+        L_self_modules_features_modules_8_modules_branch1_modules_1_modules_bn_buffers_running_mean_: torch.Tensor,
+        L_self_modules_features_modules_8_modules_branch1_modules_1_modules_bn_buffers_running_var_: torch.Tensor,
+        L_self_modules_features_modules_8_modules_branch1_modules_1_modules_bn_parameters_weight_: torch.nn.parameter.Parameter,
+        L_self_modules_features_modules_8_modules_branch1_modules_1_modules_bn_parameters_bias_: torch.nn.parameter.Parameter,
+        L_self_modules_features_modules_8_modules_branch2_modules_0_modules_conv_parameters_weight_: torch.nn.parameter.Parameter,
+        L_self_modules_features_modules_8_modules_branch2_modules_0_modules_bn_buffers_running_mean_: torch.Tensor,
+        L_self_modules_features_modules_8_modules_branch2_modules_0_modules_bn_buffers_running_var_: torch.Tensor,
+        L_self_modules_features_modules_8_modules_branch2_modules_0_modules_bn_parameters_weight_: torch.nn.parameter.Parameter,
+        L_self_modules_features_modules_8_modules_branch2_modules_0_modules_bn_parameters_bias_: torch.nn.parameter.Parameter,
+        L_self_modules_features_modules_8_modules_branch2_modules_1_modules_conv_parameters_weight_: torch.nn.parameter.Parameter,
+        L_self_modules_features_modules_8_modules_branch2_modules_1_modules_bn_buffers_running_mean_: torch.Tensor,
+        L_self_modules_features_modules_8_modules_branch2_modules_1_modules_bn_buffers_running_var_: torch.Tensor,
+        L_self_modules_features_modules_8_modules_branch2_modules_1_modules_bn_parameters_weight_: torch.nn.parameter.Parameter,
+        L_self_modules_features_modules_8_modules_branch2_modules_1_modules_bn_parameters_bias_: torch.nn.parameter.Parameter,
+        L_self_modules_features_modules_8_modules_branch2_modules_2_modules_conv_parameters_weight_: torch.nn.parameter.Parameter,
+        L_self_modules_features_modules_8_modules_branch2_modules_2_modules_bn_buffers_running_mean_: torch.Tensor,
+        L_self_modules_features_modules_8_modules_branch2_modules_2_modules_bn_buffers_running_var_: torch.Tensor,
+        L_self_modules_features_modules_8_modules_branch2_modules_2_modules_bn_parameters_weight_: torch.nn.parameter.Parameter,
+        L_self_modules_features_modules_8_modules_branch2_modules_2_modules_bn_parameters_bias_: torch.nn.parameter.Parameter,
+        L_self_modules_features_modules_8_modules_branch3_modules_1_modules_conv_parameters_weight_: torch.nn.parameter.Parameter,
+        L_self_modules_features_modules_8_modules_branch3_modules_1_modules_bn_buffers_running_mean_: torch.Tensor,
+        L_self_modules_features_modules_8_modules_branch3_modules_1_modules_bn_buffers_running_var_: torch.Tensor,
+        L_self_modules_features_modules_8_modules_branch3_modules_1_modules_bn_parameters_weight_: torch.nn.parameter.Parameter,
+        L_self_modules_features_modules_8_modules_branch3_modules_1_modules_bn_parameters_bias_: torch.nn.parameter.Parameter,
+        L_self_modules_features_modules_9_modules_branch0_modules_conv_parameters_weight_: torch.nn.parameter.Parameter,
+        L_self_modules_features_modules_9_modules_branch0_modules_bn_buffers_running_mean_: torch.Tensor,
+        L_self_modules_features_modules_9_modules_branch0_modules_bn_buffers_running_var_: torch.Tensor,
+        L_self_modules_features_modules_9_modules_branch0_modules_bn_parameters_weight_: torch.nn.parameter.Parameter,
+        L_self_modules_features_modules_9_modules_branch0_modules_bn_parameters_bias_: torch.nn.parameter.Parameter,
+        L_self_modules_features_modules_9_modules_branch1_modules_0_modules_conv_parameters_weight_: torch.nn.parameter.Parameter,
+        L_self_modules_features_modules_9_modules_branch1_modules_0_modules_bn_buffers_running_mean_: torch.Tensor,
+        L_self_modules_features_modules_9_modules_branch1_modules_0_modules_bn_buffers_running_var_: torch.Tensor,
+        L_self_modules_features_modules_9_modules_branch1_modules_0_modules_bn_parameters_weight_: torch.nn.parameter.Parameter,
+        L_self_modules_features_modules_9_modules_branch1_modules_0_modules_bn_parameters_bias_: torch.nn.parameter.Parameter,
+        L_self_modules_features_modules_9_modules_branch1_modules_1_modules_conv_parameters_weight_: torch.nn.parameter.Parameter,
+        L_self_modules_features_modules_9_modules_branch1_modules_1_modules_bn_buffers_running_mean_: torch.Tensor,
+        L_self_modules_features_modules_9_modules_branch1_modules_1_modules_bn_buffers_running_var_: torch.Tensor,
+        L_self_modules_features_modules_9_modules_branch1_modules_1_modules_bn_parameters_weight_: torch.nn.parameter.Parameter,
+        L_self_modules_features_modules_9_modules_branch1_modules_1_modules_bn_parameters_bias_: torch.nn.parameter.Parameter,
+        L_self_modules_features_modules_9_modules_branch2_modules_0_modules_conv_parameters_weight_: torch.nn.parameter.Parameter,
+        L_self_modules_features_modules_9_modules_branch2_modules_0_modules_bn_buffers_running_mean_: torch.Tensor,
+        L_self_modules_features_modules_9_modules_branch2_modules_0_modules_bn_buffers_running_var_: torch.Tensor,
+        L_self_modules_features_modules_9_modules_branch2_modules_0_modules_bn_parameters_weight_: torch.nn.parameter.Parameter,
+        L_self_modules_features_modules_9_modules_branch2_modules_0_modules_bn_parameters_bias_: torch.nn.parameter.Parameter,
+        L_self_modules_features_modules_9_modules_branch2_modules_1_modules_conv_parameters_weight_: torch.nn.parameter.Parameter,
+        L_self_modules_features_modules_9_modules_branch2_modules_1_modules_bn_buffers_running_mean_: torch.Tensor,
+        L_self_modules_features_modules_9_modules_branch2_modules_1_modules_bn_buffers_running_var_: torch.Tensor,
+        L_self_modules_features_modules_9_modules_branch2_modules_1_modules_bn_parameters_weight_: torch.nn.parameter.Parameter,
+        L_self_modules_features_modules_9_modules_branch2_modules_1_modules_bn_parameters_bias_: torch.nn.parameter.Parameter,
+        L_self_modules_features_modules_9_modules_branch2_modules_2_modules_conv_parameters_weight_: torch.nn.parameter.Parameter,
+        L_self_modules_features_modules_9_modules_branch2_modules_2_modules_bn_buffers_running_mean_: torch.Tensor,
+        L_self_modules_features_modules_9_modules_branch2_modules_2_modules_bn_buffers_running_var_: torch.Tensor,
+        L_self_modules_features_modules_9_modules_branch2_modules_2_modules_bn_parameters_weight_: torch.nn.parameter.Parameter,
+        L_self_modules_features_modules_9_modules_branch2_modules_2_modules_bn_parameters_bias_: torch.nn.parameter.Parameter,
+        L_self_modules_features_modules_9_modules_branch3_modules_1_modules_conv_parameters_weight_: torch.nn.parameter.Parameter,
+        L_self_modules_features_modules_9_modules_branch3_modules_1_modules_bn_buffers_running_mean_: torch.Tensor,
+        L_self_modules_features_modules_9_modules_branch3_modules_1_modules_bn_buffers_running_var_: torch.Tensor,
+        L_self_modules_features_modules_9_modules_branch3_modules_1_modules_bn_parameters_weight_: torch.nn.parameter.Parameter,
+        L_self_modules_features_modules_9_modules_branch3_modules_1_modules_bn_parameters_bias_: torch.nn.parameter.Parameter,
+        L_self_modules_features_modules_10_modules_branch0_modules_conv_parameters_weight_: torch.nn.parameter.Parameter,
+        L_self_modules_features_modules_10_modules_branch0_modules_bn_buffers_running_mean_: torch.Tensor,
+        L_self_modules_features_modules_10_modules_branch0_modules_bn_buffers_running_var_: torch.Tensor,
+        L_self_modules_features_modules_10_modules_branch0_modules_bn_parameters_weight_: torch.nn.parameter.Parameter,
+        L_self_modules_features_modules_10_modules_branch0_modules_bn_parameters_bias_: torch.nn.parameter.Parameter,
+        L_self_modules_features_modules_10_modules_branch1_modules_0_modules_conv_parameters_weight_: torch.nn.parameter.Parameter,
+        L_self_modules_features_modules_10_modules_branch1_modules_0_modules_bn_buffers_running_mean_: torch.Tensor,
+        L_self_modules_features_modules_10_modules_branch1_modules_0_modules_bn_buffers_running_var_: torch.Tensor,
+        L_self_modules_features_modules_10_modules_branch1_modules_0_modules_bn_parameters_weight_: torch.nn.parameter.Parameter,
+        L_self_modules_features_modules_10_modules_branch1_modules_0_modules_bn_parameters_bias_: torch.nn.parameter.Parameter,
+        L_self_modules_features_modules_10_modules_branch1_modules_1_modules_conv_parameters_weight_: torch.nn.parameter.Parameter,
+        L_self_modules_features_modules_10_modules_branch1_modules_1_modules_bn_buffers_running_mean_: torch.Tensor,
+        L_self_modules_features_modules_10_modules_branch1_modules_1_modules_bn_buffers_running_var_: torch.Tensor,
+        L_self_modules_features_modules_10_modules_branch1_modules_1_modules_bn_parameters_weight_: torch.nn.parameter.Parameter,
+        L_self_modules_features_modules_10_modules_branch1_modules_1_modules_bn_parameters_bias_: torch.nn.parameter.Parameter,
+        L_self_modules_features_modules_10_modules_branch1_modules_2_modules_conv_parameters_weight_: torch.nn.parameter.Parameter,
+        L_self_modules_features_modules_10_modules_branch1_modules_2_modules_bn_buffers_running_mean_: torch.Tensor,
+        L_self_modules_features_modules_10_modules_branch1_modules_2_modules_bn_buffers_running_var_: torch.Tensor,
+        L_self_modules_features_modules_10_modules_branch1_modules_2_modules_bn_parameters_weight_: torch.nn.parameter.Parameter,
+        L_self_modules_features_modules_10_modules_branch1_modules_2_modules_bn_parameters_bias_: torch.nn.parameter.Parameter,
+        L_self_modules_features_modules_11_modules_branch0_modules_conv_parameters_weight_: torch.nn.parameter.Parameter,
+        L_self_modules_features_modules_11_modules_branch0_modules_bn_buffers_running_mean_: torch.Tensor,
+        L_self_modules_features_modules_11_modules_branch0_modules_bn_buffers_running_var_: torch.Tensor,
+        L_self_modules_features_modules_11_modules_branch0_modules_bn_parameters_weight_: torch.nn.parameter.Parameter,
+        L_self_modules_features_modules_11_modules_branch0_modules_bn_parameters_bias_: torch.nn.parameter.Parameter,
+        L_self_modules_features_modules_11_modules_branch1_modules_0_modules_conv_parameters_weight_: torch.nn.parameter.Parameter,
+        L_self_modules_features_modules_11_modules_branch1_modules_0_modules_bn_buffers_running_mean_: torch.Tensor,
+        L_self_modules_features_modules_11_modules_branch1_modules_0_modules_bn_buffers_running_var_: torch.Tensor,
+        L_self_modules_features_modules_11_modules_branch1_modules_0_modules_bn_parameters_weight_: torch.nn.parameter.Parameter,
+        L_self_modules_features_modules_11_modules_branch1_modules_0_modules_bn_parameters_bias_: torch.nn.parameter.Parameter,
+        L_self_modules_features_modules_11_modules_branch1_modules_1_modules_conv_parameters_weight_: torch.nn.parameter.Parameter,
+        L_self_modules_features_modules_11_modules_branch1_modules_1_modules_bn_buffers_running_mean_: torch.Tensor,
+        L_self_modules_features_modules_11_modules_branch1_modules_1_modules_bn_buffers_running_var_: torch.Tensor,
+        L_self_modules_features_modules_11_modules_branch1_modules_1_modules_bn_parameters_weight_: torch.nn.parameter.Parameter,
+        L_self_modules_features_modules_11_modules_branch1_modules_1_modules_bn_parameters_bias_: torch.nn.parameter.Parameter,
+        L_self_modules_features_modules_11_modules_branch1_modules_2_modules_conv_parameters_weight_: torch.nn.parameter.Parameter,
+        L_self_modules_features_modules_11_modules_branch1_modules_2_modules_bn_buffers_running_mean_: torch.Tensor,
+        L_self_modules_features_modules_11_modules_branch1_modules_2_modules_bn_buffers_running_var_: torch.Tensor,
+        L_self_modules_features_modules_11_modules_branch1_modules_2_modules_bn_parameters_weight_: torch.nn.parameter.Parameter,
+        L_self_modules_features_modules_11_modules_branch1_modules_2_modules_bn_parameters_bias_: torch.nn.parameter.Parameter,
+        L_self_modules_features_modules_11_modules_branch2_modules_0_modules_conv_parameters_weight_: torch.nn.parameter.Parameter,
+        L_self_modules_features_modules_11_modules_branch2_modules_0_modules_bn_buffers_running_mean_: torch.Tensor,
+        L_self_modules_features_modules_11_modules_branch2_modules_0_modules_bn_buffers_running_var_: torch.Tensor,
+        L_self_modules_features_modules_11_modules_branch2_modules_0_modules_bn_parameters_weight_: torch.nn.parameter.Parameter,
+        L_self_modules_features_modules_11_modules_branch2_modules_0_modules_bn_parameters_bias_: torch.nn.parameter.Parameter,
+        L_self_modules_features_modules_11_modules_branch2_modules_1_modules_conv_parameters_weight_: torch.nn.parameter.Parameter,
+        L_self_modules_features_modules_11_modules_branch2_modules_1_modules_bn_buffers_running_mean_: torch.Tensor,
+        L_self_modules_features_modules_11_modules_branch2_modules_1_modules_bn_buffers_running_var_: torch.Tensor,
+        L_self_modules_features_modules_11_modules_branch2_modules_1_modules_bn_parameters_weight_: torch.nn.parameter.Parameter,
+        L_self_modules_features_modules_11_modules_branch2_modules_1_modules_bn_parameters_bias_: torch.nn.parameter.Parameter,
+        L_self_modules_features_modules_11_modules_branch2_modules_2_modules_conv_parameters_weight_: torch.nn.parameter.Parameter,
+        L_self_modules_features_modules_11_modules_branch2_modules_2_modules_bn_buffers_running_mean_: torch.Tensor,
+        L_self_modules_features_modules_11_modules_branch2_modules_2_modules_bn_buffers_running_var_: torch.Tensor,
+        L_self_modules_features_modules_11_modules_branch2_modules_2_modules_bn_parameters_weight_: torch.nn.parameter.Parameter,
+        L_self_modules_features_modules_11_modules_branch2_modules_2_modules_bn_parameters_bias_: torch.nn.parameter.Parameter,
+        L_self_modules_features_modules_11_modules_branch2_modules_3_modules_conv_parameters_weight_: torch.nn.parameter.Parameter,
+        L_self_modules_features_modules_11_modules_branch2_modules_3_modules_bn_buffers_running_mean_: torch.Tensor,
+        L_self_modules_features_modules_11_modules_branch2_modules_3_modules_bn_buffers_running_var_: torch.Tensor,
+        L_self_modules_features_modules_11_modules_branch2_modules_3_modules_bn_parameters_weight_: torch.nn.parameter.Parameter,
+        L_self_modules_features_modules_11_modules_branch2_modules_3_modules_bn_parameters_bias_: torch.nn.parameter.Parameter,
+        L_self_modules_features_modules_11_modules_branch2_modules_4_modules_conv_parameters_weight_: torch.nn.parameter.Parameter,
+        L_self_modules_features_modules_11_modules_branch2_modules_4_modules_bn_buffers_running_mean_: torch.Tensor,
+        L_self_modules_features_modules_11_modules_branch2_modules_4_modules_bn_buffers_running_var_: torch.Tensor,
+        L_self_modules_features_modules_11_modules_branch2_modules_4_modules_bn_parameters_weight_: torch.nn.parameter.Parameter,
+        L_self_modules_features_modules_11_modules_branch2_modules_4_modules_bn_parameters_bias_: torch.nn.parameter.Parameter,
+        L_self_modules_features_modules_11_modules_branch3_modules_1_modules_conv_parameters_weight_: torch.nn.parameter.Parameter,
+        L_self_modules_features_modules_11_modules_branch3_modules_1_modules_bn_buffers_running_mean_: torch.Tensor,
+        L_self_modules_features_modules_11_modules_branch3_modules_1_modules_bn_buffers_running_var_: torch.Tensor,
+        L_self_modules_features_modules_11_modules_branch3_modules_1_modules_bn_parameters_weight_: torch.nn.parameter.Parameter,
+        L_self_modules_features_modules_11_modules_branch3_modules_1_modules_bn_parameters_bias_: torch.nn.parameter.Parameter,
+        L_self_modules_features_modules_12_modules_branch0_modules_conv_parameters_weight_: torch.nn.parameter.Parameter,
+        L_self_modules_features_modules_12_modules_branch0_modules_bn_buffers_running_mean_: torch.Tensor,
+        L_self_modules_features_modules_12_modules_branch0_modules_bn_buffers_running_var_: torch.Tensor,
+        L_self_modules_features_modules_12_modules_branch0_modules_bn_parameters_weight_: torch.nn.parameter.Parameter,
+        L_self_modules_features_modules_12_modules_branch0_modules_bn_parameters_bias_: torch.nn.parameter.Parameter,
+        L_self_modules_features_modules_12_modules_branch1_modules_0_modules_conv_parameters_weight_: torch.nn.parameter.Parameter,
+        L_self_modules_features_modules_12_modules_branch1_modules_0_modules_bn_buffers_running_mean_: torch.Tensor,
+        L_self_modules_features_modules_12_modules_branch1_modules_0_modules_bn_buffers_running_var_: torch.Tensor,
+        L_self_modules_features_modules_12_modules_branch1_modules_0_modules_bn_parameters_weight_: torch.nn.parameter.Parameter,
+        L_self_modules_features_modules_12_modules_branch1_modules_0_modules_bn_parameters_bias_: torch.nn.parameter.Parameter,
+        L_self_modules_features_modules_12_modules_branch1_modules_1_modules_conv_parameters_weight_: torch.nn.parameter.Parameter,
+        L_self_modules_features_modules_12_modules_branch1_modules_1_modules_bn_buffers_running_mean_: torch.Tensor,
+        L_self_modules_features_modules_12_modules_branch1_modules_1_modules_bn_buffers_running_var_: torch.Tensor,
+        L_self_modules_features_modules_12_modules_branch1_modules_1_modules_bn_parameters_weight_: torch.nn.parameter.Parameter,
+        L_self_modules_features_modules_12_modules_branch1_modules_1_modules_bn_parameters_bias_: torch.nn.parameter.Parameter,
+        L_self_modules_features_modules_12_modules_branch1_modules_2_modules_conv_parameters_weight_: torch.nn.parameter.Parameter,
+        L_self_modules_features_modules_12_modules_branch1_modules_2_modules_bn_buffers_running_mean_: torch.Tensor,
+        L_self_modules_features_modules_12_modules_branch1_modules_2_modules_bn_buffers_running_var_: torch.Tensor,
+        L_self_modules_features_modules_12_modules_branch1_modules_2_modules_bn_parameters_weight_: torch.nn.parameter.Parameter,
+        L_self_modules_features_modules_12_modules_branch1_modules_2_modules_bn_parameters_bias_: torch.nn.parameter.Parameter,
+        L_self_modules_features_modules_12_modules_branch2_modules_0_modules_conv_parameters_weight_: torch.nn.parameter.Parameter,
+        L_self_modules_features_modules_12_modules_branch2_modules_0_modules_bn_buffers_running_mean_: torch.Tensor,
+        L_self_modules_features_modules_12_modules_branch2_modules_0_modules_bn_buffers_running_var_: torch.Tensor,
+        L_self_modules_features_modules_12_modules_branch2_modules_0_modules_bn_parameters_weight_: torch.nn.parameter.Parameter,
+        L_self_modules_features_modules_12_modules_branch2_modules_0_modules_bn_parameters_bias_: torch.nn.parameter.Parameter,
+        L_self_modules_features_modules_12_modules_branch2_modules_1_modules_conv_parameters_weight_: torch.nn.parameter.Parameter,
+        L_self_modules_features_modules_12_modules_branch2_modules_1_modules_bn_buffers_running_mean_: torch.Tensor,
+        L_self_modules_features_modules_12_modules_branch2_modules_1_modules_bn_buffers_running_var_: torch.Tensor,
+        L_self_modules_features_modules_12_modules_branch2_modules_1_modules_bn_parameters_weight_: torch.nn.parameter.Parameter,
+        L_self_modules_features_modules_12_modules_branch2_modules_1_modules_bn_parameters_bias_: torch.nn.parameter.Parameter,
+        L_self_modules_features_modules_12_modules_branch2_modules_2_modules_conv_parameters_weight_: torch.nn.parameter.Parameter,
+        L_self_modules_features_modules_12_modules_branch2_modules_2_modules_bn_buffers_running_mean_: torch.Tensor,
+        L_self_modules_features_modules_12_modules_branch2_modules_2_modules_bn_buffers_running_var_: torch.Tensor,
+        L_self_modules_features_modules_12_modules_branch2_modules_2_modules_bn_parameters_weight_: torch.nn.parameter.Parameter,
+        L_self_modules_features_modules_12_modules_branch2_modules_2_modules_bn_parameters_bias_: torch.nn.parameter.Parameter,
+        L_self_modules_features_modules_12_modules_branch2_modules_3_modules_conv_parameters_weight_: torch.nn.parameter.Parameter,
+        L_self_modules_features_modules_12_modules_branch2_modules_3_modules_bn_buffers_running_mean_: torch.Tensor,
+        L_self_modules_features_modules_12_modules_branch2_modules_3_modules_bn_buffers_running_var_: torch.Tensor,
+        L_self_modules_features_modules_12_modules_branch2_modules_3_modules_bn_parameters_weight_: torch.nn.parameter.Parameter,
+        L_self_modules_features_modules_12_modules_branch2_modules_3_modules_bn_parameters_bias_: torch.nn.parameter.Parameter,
+        L_self_modules_features_modules_12_modules_branch2_modules_4_modules_conv_parameters_weight_: torch.nn.parameter.Parameter,
+        L_self_modules_features_modules_12_modules_branch2_modules_4_modules_bn_buffers_running_mean_: torch.Tensor,
+        L_self_modules_features_modules_12_modules_branch2_modules_4_modules_bn_buffers_running_var_: torch.Tensor,
+        L_self_modules_features_modules_12_modules_branch2_modules_4_modules_bn_parameters_weight_: torch.nn.parameter.Parameter,
+        L_self_modules_features_modules_12_modules_branch2_modules_4_modules_bn_parameters_bias_: torch.nn.parameter.Parameter,
+        L_self_modules_features_modules_12_modules_branch3_modules_1_modules_conv_parameters_weight_: torch.nn.parameter.Parameter,
+        L_self_modules_features_modules_12_modules_branch3_modules_1_modules_bn_buffers_running_mean_: torch.Tensor,
+        L_self_modules_features_modules_12_modules_branch3_modules_1_modules_bn_buffers_running_var_: torch.Tensor,
+        L_self_modules_features_modules_12_modules_branch3_modules_1_modules_bn_parameters_weight_: torch.nn.parameter.Parameter,
+        L_self_modules_features_modules_12_modules_branch3_modules_1_modules_bn_parameters_bias_: torch.nn.parameter.Parameter,
+        L_self_modules_features_modules_13_modules_branch0_modules_conv_parameters_weight_: torch.nn.parameter.Parameter,
+        L_self_modules_features_modules_13_modules_branch0_modules_bn_buffers_running_mean_: torch.Tensor,
+        L_self_modules_features_modules_13_modules_branch0_modules_bn_buffers_running_var_: torch.Tensor,
+        L_self_modules_features_modules_13_modules_branch0_modules_bn_parameters_weight_: torch.nn.parameter.Parameter,
+        L_self_modules_features_modules_13_modules_branch0_modules_bn_parameters_bias_: torch.nn.parameter.Parameter,
+        L_self_modules_features_modules_13_modules_branch1_modules_0_modules_conv_parameters_weight_: torch.nn.parameter.Parameter,
+        L_self_modules_features_modules_13_modules_branch1_modules_0_modules_bn_buffers_running_mean_: torch.Tensor,
+        L_self_modules_features_modules_13_modules_branch1_modules_0_modules_bn_buffers_running_var_: torch.Tensor,
+        L_self_modules_features_modules_13_modules_branch1_modules_0_modules_bn_parameters_weight_: torch.nn.parameter.Parameter,
+        L_self_modules_features_modules_13_modules_branch1_modules_0_modules_bn_parameters_bias_: torch.nn.parameter.Parameter,
+        L_self_modules_features_modules_13_modules_branch1_modules_1_modules_conv_parameters_weight_: torch.nn.parameter.Parameter,
+        L_self_modules_features_modules_13_modules_branch1_modules_1_modules_bn_buffers_running_mean_: torch.Tensor,
+        L_self_modules_features_modules_13_modules_branch1_modules_1_modules_bn_buffers_running_var_: torch.Tensor,
+        L_self_modules_features_modules_13_modules_branch1_modules_1_modules_bn_parameters_weight_: torch.nn.parameter.Parameter,
+        L_self_modules_features_modules_13_modules_branch1_modules_1_modules_bn_parameters_bias_: torch.nn.parameter.Parameter,
+        L_self_modules_features_modules_13_modules_branch1_modules_2_modules_conv_parameters_weight_: torch.nn.parameter.Parameter,
+        L_self_modules_features_modules_13_modules_branch1_modules_2_modules_bn_buffers_running_mean_: torch.Tensor,
+        L_self_modules_features_modules_13_modules_branch1_modules_2_modules_bn_buffers_running_var_: torch.Tensor,
+        L_self_modules_features_modules_13_modules_branch1_modules_2_modules_bn_parameters_weight_: torch.nn.parameter.Parameter,
+        L_self_modules_features_modules_13_modules_branch1_modules_2_modules_bn_parameters_bias_: torch.nn.parameter.Parameter,
+        L_self_modules_features_modules_13_modules_branch2_modules_0_modules_conv_parameters_weight_: torch.nn.parameter.Parameter,
+        L_self_modules_features_modules_13_modules_branch2_modules_0_modules_bn_buffers_running_mean_: torch.Tensor,
+        L_self_modules_features_modules_13_modules_branch2_modules_0_modules_bn_buffers_running_var_: torch.Tensor,
+        L_self_modules_features_modules_13_modules_branch2_modules_0_modules_bn_parameters_weight_: torch.nn.parameter.Parameter,
+        L_self_modules_features_modules_13_modules_branch2_modules_0_modules_bn_parameters_bias_: torch.nn.parameter.Parameter,
+        L_self_modules_features_modules_13_modules_branch2_modules_1_modules_conv_parameters_weight_: torch.nn.parameter.Parameter,
+        L_self_modules_features_modules_13_modules_branch2_modules_1_modules_bn_buffers_running_mean_: torch.Tensor,
+        L_self_modules_features_modules_13_modules_branch2_modules_1_modules_bn_buffers_running_var_: torch.Tensor,
+        L_self_modules_features_modules_13_modules_branch2_modules_1_modules_bn_parameters_weight_: torch.nn.parameter.Parameter,
+        L_self_modules_features_modules_13_modules_branch2_modules_1_modules_bn_parameters_bias_: torch.nn.parameter.Parameter,
+        L_self_modules_features_modules_13_modules_branch2_modules_2_modules_conv_parameters_weight_: torch.nn.parameter.Parameter,
+        L_self_modules_features_modules_13_modules_branch2_modules_2_modules_bn_buffers_running_mean_: torch.Tensor,
+        L_self_modules_features_modules_13_modules_branch2_modules_2_modules_bn_buffers_running_var_: torch.Tensor,
+        L_self_modules_features_modules_13_modules_branch2_modules_2_modules_bn_parameters_weight_: torch.nn.parameter.Parameter,
+        L_self_modules_features_modules_13_modules_branch2_modules_2_modules_bn_parameters_bias_: torch.nn.parameter.Parameter,
+        L_self_modules_features_modules_13_modules_branch2_modules_3_modules_conv_parameters_weight_: torch.nn.parameter.Parameter,
+        L_self_modules_features_modules_13_modules_branch2_modules_3_modules_bn_buffers_running_mean_: torch.Tensor,
+        L_self_modules_features_modules_13_modules_branch2_modules_3_modules_bn_buffers_running_var_: torch.Tensor,
+        L_self_modules_features_modules_13_modules_branch2_modules_3_modules_bn_parameters_weight_: torch.nn.parameter.Parameter,
+        L_self_modules_features_modules_13_modules_branch2_modules_3_modules_bn_parameters_bias_: torch.nn.parameter.Parameter,
+        L_self_modules_features_modules_13_modules_branch2_modules_4_modules_conv_parameters_weight_: torch.nn.parameter.Parameter,
+        L_self_modules_features_modules_13_modules_branch2_modules_4_modules_bn_buffers_running_mean_: torch.Tensor,
+        L_self_modules_features_modules_13_modules_branch2_modules_4_modules_bn_buffers_running_var_: torch.Tensor,
+        L_self_modules_features_modules_13_modules_branch2_modules_4_modules_bn_parameters_weight_: torch.nn.parameter.Parameter,
+        L_self_modules_features_modules_13_modules_branch2_modules_4_modules_bn_parameters_bias_: torch.nn.parameter.Parameter,
+        L_self_modules_features_modules_13_modules_branch3_modules_1_modules_conv_parameters_weight_: torch.nn.parameter.Parameter,
+        L_self_modules_features_modules_13_modules_branch3_modules_1_modules_bn_buffers_running_mean_: torch.Tensor,
+        L_self_modules_features_modules_13_modules_branch3_modules_1_modules_bn_buffers_running_var_: torch.Tensor,
+        L_self_modules_features_modules_13_modules_branch3_modules_1_modules_bn_parameters_weight_: torch.nn.parameter.Parameter,
+        L_self_modules_features_modules_13_modules_branch3_modules_1_modules_bn_parameters_bias_: torch.nn.parameter.Parameter,
+        L_self_modules_features_modules_14_modules_branch0_modules_conv_parameters_weight_: torch.nn.parameter.Parameter,
+        L_self_modules_features_modules_14_modules_branch0_modules_bn_buffers_running_mean_: torch.Tensor,
+        L_self_modules_features_modules_14_modules_branch0_modules_bn_buffers_running_var_: torch.Tensor,
+        L_self_modules_features_modules_14_modules_branch0_modules_bn_parameters_weight_: torch.nn.parameter.Parameter,
+        L_self_modules_features_modules_14_modules_branch0_modules_bn_parameters_bias_: torch.nn.parameter.Parameter,
+        L_self_modules_features_modules_14_modules_branch1_modules_0_modules_conv_parameters_weight_: torch.nn.parameter.Parameter,
+        L_self_modules_features_modules_14_modules_branch1_modules_0_modules_bn_buffers_running_mean_: torch.Tensor,
+        L_self_modules_features_modules_14_modules_branch1_modules_0_modules_bn_buffers_running_var_: torch.Tensor,
+        L_self_modules_features_modules_14_modules_branch1_modules_0_modules_bn_parameters_weight_: torch.nn.parameter.Parameter,
+        L_self_modules_features_modules_14_modules_branch1_modules_0_modules_bn_parameters_bias_: torch.nn.parameter.Parameter,
+        L_self_modules_features_modules_14_modules_branch1_modules_1_modules_conv_parameters_weight_: torch.nn.parameter.Parameter,
+        L_self_modules_features_modules_14_modules_branch1_modules_1_modules_bn_buffers_running_mean_: torch.Tensor,
+        L_self_modules_features_modules_14_modules_branch1_modules_1_modules_bn_buffers_running_var_: torch.Tensor,
+        L_self_modules_features_modules_14_modules_branch1_modules_1_modules_bn_parameters_weight_: torch.nn.parameter.Parameter,
+        L_self_modules_features_modules_14_modules_branch1_modules_1_modules_bn_parameters_bias_: torch.nn.parameter.Parameter,
+        L_self_modules_features_modules_14_modules_branch1_modules_2_modules_conv_parameters_weight_: torch.nn.parameter.Parameter,
+        L_self_modules_features_modules_14_modules_branch1_modules_2_modules_bn_buffers_running_mean_: torch.Tensor,
+        L_self_modules_features_modules_14_modules_branch1_modules_2_modules_bn_buffers_running_var_: torch.Tensor,
+        L_self_modules_features_modules_14_modules_branch1_modules_2_modules_bn_parameters_weight_: torch.nn.parameter.Parameter,
+        L_self_modules_features_modules_14_modules_branch1_modules_2_modules_bn_parameters_bias_: torch.nn.parameter.Parameter,
+        L_self_modules_features_modules_14_modules_branch2_modules_0_modules_conv_parameters_weight_: torch.nn.parameter.Parameter,
+        L_self_modules_features_modules_14_modules_branch2_modules_0_modules_bn_buffers_running_mean_: torch.Tensor,
+        L_self_modules_features_modules_14_modules_branch2_modules_0_modules_bn_buffers_running_var_: torch.Tensor,
+        L_self_modules_features_modules_14_modules_branch2_modules_0_modules_bn_parameters_weight_: torch.nn.parameter.Parameter,
+        L_self_modules_features_modules_14_modules_branch2_modules_0_modules_bn_parameters_bias_: torch.nn.parameter.Parameter,
+        L_self_modules_features_modules_14_modules_branch2_modules_1_modules_conv_parameters_weight_: torch.nn.parameter.Parameter,
+        L_self_modules_features_modules_14_modules_branch2_modules_1_modules_bn_buffers_running_mean_: torch.Tensor,
+        L_self_modules_features_modules_14_modules_branch2_modules_1_modules_bn_buffers_running_var_: torch.Tensor,
+        L_self_modules_features_modules_14_modules_branch2_modules_1_modules_bn_parameters_weight_: torch.nn.parameter.Parameter,
+        L_self_modules_features_modules_14_modules_branch2_modules_1_modules_bn_parameters_bias_: torch.nn.parameter.Parameter,
+        L_self_modules_features_modules_14_modules_branch2_modules_2_modules_conv_parameters_weight_: torch.nn.parameter.Parameter,
+        L_self_modules_features_modules_14_modules_branch2_modules_2_modules_bn_buffers_running_mean_: torch.Tensor,
+        L_self_modules_features_modules_14_modules_branch2_modules_2_modules_bn_buffers_running_var_: torch.Tensor,
+        L_self_modules_features_modules_14_modules_branch2_modules_2_modules_bn_parameters_weight_: torch.nn.parameter.Parameter,
+        L_self_modules_features_modules_14_modules_branch2_modules_2_modules_bn_parameters_bias_: torch.nn.parameter.Parameter,
+        L_self_modules_features_modules_14_modules_branch2_modules_3_modules_conv_parameters_weight_: torch.nn.parameter.Parameter,
+        L_self_modules_features_modules_14_modules_branch2_modules_3_modules_bn_buffers_running_mean_: torch.Tensor,
+        L_self_modules_features_modules_14_modules_branch2_modules_3_modules_bn_buffers_running_var_: torch.Tensor,
+        L_self_modules_features_modules_14_modules_branch2_modules_3_modules_bn_parameters_weight_: torch.nn.parameter.Parameter,
+        L_self_modules_features_modules_14_modules_branch2_modules_3_modules_bn_parameters_bias_: torch.nn.parameter.Parameter,
+        L_self_modules_features_modules_14_modules_branch2_modules_4_modules_conv_parameters_weight_: torch.nn.parameter.Parameter,
+        L_self_modules_features_modules_14_modules_branch2_modules_4_modules_bn_buffers_running_mean_: torch.Tensor,
+        L_self_modules_features_modules_14_modules_branch2_modules_4_modules_bn_buffers_running_var_: torch.Tensor,
+        L_self_modules_features_modules_14_modules_branch2_modules_4_modules_bn_parameters_weight_: torch.nn.parameter.Parameter,
+        L_self_modules_features_modules_14_modules_branch2_modules_4_modules_bn_parameters_bias_: torch.nn.parameter.Parameter,
+        L_self_modules_features_modules_14_modules_branch3_modules_1_modules_conv_parameters_weight_: torch.nn.parameter.Parameter,
+        L_self_modules_features_modules_14_modules_branch3_modules_1_modules_bn_buffers_running_mean_: torch.Tensor,
+        L_self_modules_features_modules_14_modules_branch3_modules_1_modules_bn_buffers_running_var_: torch.Tensor,
+        L_self_modules_features_modules_14_modules_branch3_modules_1_modules_bn_parameters_weight_: torch.nn.parameter.Parameter,
+        L_self_modules_features_modules_14_modules_branch3_modules_1_modules_bn_parameters_bias_: torch.nn.parameter.Parameter,
+        L_self_modules_features_modules_15_modules_branch0_modules_conv_parameters_weight_: torch.nn.parameter.Parameter,
+        L_self_modules_features_modules_15_modules_branch0_modules_bn_buffers_running_mean_: torch.Tensor,
+        L_self_modules_features_modules_15_modules_branch0_modules_bn_buffers_running_var_: torch.Tensor,
+        L_self_modules_features_modules_15_modules_branch0_modules_bn_parameters_weight_: torch.nn.parameter.Parameter,
+        L_self_modules_features_modules_15_modules_branch0_modules_bn_parameters_bias_: torch.nn.parameter.Parameter,
+        L_self_modules_features_modules_15_modules_branch1_modules_0_modules_conv_parameters_weight_: torch.nn.parameter.Parameter,
+        L_self_modules_features_modules_15_modules_branch1_modules_0_modules_bn_buffers_running_mean_: torch.Tensor,
+        L_self_modules_features_modules_15_modules_branch1_modules_0_modules_bn_buffers_running_var_: torch.Tensor,
+        L_self_modules_features_modules_15_modules_branch1_modules_0_modules_bn_parameters_weight_: torch.nn.parameter.Parameter,
+        L_self_modules_features_modules_15_modules_branch1_modules_0_modules_bn_parameters_bias_: torch.nn.parameter.Parameter,
+        L_self_modules_features_modules_15_modules_branch1_modules_1_modules_conv_parameters_weight_: torch.nn.parameter.Parameter,
+        L_self_modules_features_modules_15_modules_branch1_modules_1_modules_bn_buffers_running_mean_: torch.Tensor,
+        L_self_modules_features_modules_15_modules_branch1_modules_1_modules_bn_buffers_running_var_: torch.Tensor,
+        L_self_modules_features_modules_15_modules_branch1_modules_1_modules_bn_parameters_weight_: torch.nn.parameter.Parameter,
+        L_self_modules_features_modules_15_modules_branch1_modules_1_modules_bn_parameters_bias_: torch.nn.parameter.Parameter,
+        L_self_modules_features_modules_15_modules_branch1_modules_2_modules_conv_parameters_weight_: torch.nn.parameter.Parameter,
+        L_self_modules_features_modules_15_modules_branch1_modules_2_modules_bn_buffers_running_mean_: torch.Tensor,
+        L_self_modules_features_modules_15_modules_branch1_modules_2_modules_bn_buffers_running_var_: torch.Tensor,
+        L_self_modules_features_modules_15_modules_branch1_modules_2_modules_bn_parameters_weight_: torch.nn.parameter.Parameter,
+        L_self_modules_features_modules_15_modules_branch1_modules_2_modules_bn_parameters_bias_: torch.nn.parameter.Parameter,
+        L_self_modules_features_modules_15_modules_branch2_modules_0_modules_conv_parameters_weight_: torch.nn.parameter.Parameter,
+        L_self_modules_features_modules_15_modules_branch2_modules_0_modules_bn_buffers_running_mean_: torch.Tensor,
+        L_self_modules_features_modules_15_modules_branch2_modules_0_modules_bn_buffers_running_var_: torch.Tensor,
+        L_self_modules_features_modules_15_modules_branch2_modules_0_modules_bn_parameters_weight_: torch.nn.parameter.Parameter,
+        L_self_modules_features_modules_15_modules_branch2_modules_0_modules_bn_parameters_bias_: torch.nn.parameter.Parameter,
+        L_self_modules_features_modules_15_modules_branch2_modules_1_modules_conv_parameters_weight_: torch.nn.parameter.Parameter,
+        L_self_modules_features_modules_15_modules_branch2_modules_1_modules_bn_buffers_running_mean_: torch.Tensor,
+        L_self_modules_features_modules_15_modules_branch2_modules_1_modules_bn_buffers_running_var_: torch.Tensor,
+        L_self_modules_features_modules_15_modules_branch2_modules_1_modules_bn_parameters_weight_: torch.nn.parameter.Parameter,
+        L_self_modules_features_modules_15_modules_branch2_modules_1_modules_bn_parameters_bias_: torch.nn.parameter.Parameter,
+        L_self_modules_features_modules_15_modules_branch2_modules_2_modules_conv_parameters_weight_: torch.nn.parameter.Parameter,
+        L_self_modules_features_modules_15_modules_branch2_modules_2_modules_bn_buffers_running_mean_: torch.Tensor,
+        L_self_modules_features_modules_15_modules_branch2_modules_2_modules_bn_buffers_running_var_: torch.Tensor,
+        L_self_modules_features_modules_15_modules_branch2_modules_2_modules_bn_parameters_weight_: torch.nn.parameter.Parameter,
+        L_self_modules_features_modules_15_modules_branch2_modules_2_modules_bn_parameters_bias_: torch.nn.parameter.Parameter,
+        L_self_modules_features_modules_15_modules_branch2_modules_3_modules_conv_parameters_weight_: torch.nn.parameter.Parameter,
+        L_self_modules_features_modules_15_modules_branch2_modules_3_modules_bn_buffers_running_mean_: torch.Tensor,
+        L_self_modules_features_modules_15_modules_branch2_modules_3_modules_bn_buffers_running_var_: torch.Tensor,
+        L_self_modules_features_modules_15_modules_branch2_modules_3_modules_bn_parameters_weight_: torch.nn.parameter.Parameter,
+        L_self_modules_features_modules_15_modules_branch2_modules_3_modules_bn_parameters_bias_: torch.nn.parameter.Parameter,
+        L_self_modules_features_modules_15_modules_branch2_modules_4_modules_conv_parameters_weight_: torch.nn.parameter.Parameter,
+        L_self_modules_features_modules_15_modules_branch2_modules_4_modules_bn_buffers_running_mean_: torch.Tensor,
+        L_self_modules_features_modules_15_modules_branch2_modules_4_modules_bn_buffers_running_var_: torch.Tensor,
+        L_self_modules_features_modules_15_modules_branch2_modules_4_modules_bn_parameters_weight_: torch.nn.parameter.Parameter,
+        L_self_modules_features_modules_15_modules_branch2_modules_4_modules_bn_parameters_bias_: torch.nn.parameter.Parameter,
+        L_self_modules_features_modules_15_modules_branch3_modules_1_modules_conv_parameters_weight_: torch.nn.parameter.Parameter,
+        L_self_modules_features_modules_15_modules_branch3_modules_1_modules_bn_buffers_running_mean_: torch.Tensor,
+        L_self_modules_features_modules_15_modules_branch3_modules_1_modules_bn_buffers_running_var_: torch.Tensor,
+        L_self_modules_features_modules_15_modules_branch3_modules_1_modules_bn_parameters_weight_: torch.nn.parameter.Parameter,
+        L_self_modules_features_modules_15_modules_branch3_modules_1_modules_bn_parameters_bias_: torch.nn.parameter.Parameter,
+        L_self_modules_features_modules_16_modules_branch0_modules_conv_parameters_weight_: torch.nn.parameter.Parameter,
+        L_self_modules_features_modules_16_modules_branch0_modules_bn_buffers_running_mean_: torch.Tensor,
+        L_self_modules_features_modules_16_modules_branch0_modules_bn_buffers_running_var_: torch.Tensor,
+        L_self_modules_features_modules_16_modules_branch0_modules_bn_parameters_weight_: torch.nn.parameter.Parameter,
+        L_self_modules_features_modules_16_modules_branch0_modules_bn_parameters_bias_: torch.nn.parameter.Parameter,
+        L_self_modules_features_modules_16_modules_branch1_modules_0_modules_conv_parameters_weight_: torch.nn.parameter.Parameter,
+        L_self_modules_features_modules_16_modules_branch1_modules_0_modules_bn_buffers_running_mean_: torch.Tensor,
+        L_self_modules_features_modules_16_modules_branch1_modules_0_modules_bn_buffers_running_var_: torch.Tensor,
+        L_self_modules_features_modules_16_modules_branch1_modules_0_modules_bn_parameters_weight_: torch.nn.parameter.Parameter,
+        L_self_modules_features_modules_16_modules_branch1_modules_0_modules_bn_parameters_bias_: torch.nn.parameter.Parameter,
+        L_self_modules_features_modules_16_modules_branch1_modules_1_modules_conv_parameters_weight_: torch.nn.parameter.Parameter,
+        L_self_modules_features_modules_16_modules_branch1_modules_1_modules_bn_buffers_running_mean_: torch.Tensor,
+        L_self_modules_features_modules_16_modules_branch1_modules_1_modules_bn_buffers_running_var_: torch.Tensor,
+        L_self_modules_features_modules_16_modules_branch1_modules_1_modules_bn_parameters_weight_: torch.nn.parameter.Parameter,
+        L_self_modules_features_modules_16_modules_branch1_modules_1_modules_bn_parameters_bias_: torch.nn.parameter.Parameter,
+        L_self_modules_features_modules_16_modules_branch1_modules_2_modules_conv_parameters_weight_: torch.nn.parameter.Parameter,
+        L_self_modules_features_modules_16_modules_branch1_modules_2_modules_bn_buffers_running_mean_: torch.Tensor,
+        L_self_modules_features_modules_16_modules_branch1_modules_2_modules_bn_buffers_running_var_: torch.Tensor,
+        L_self_modules_features_modules_16_modules_branch1_modules_2_modules_bn_parameters_weight_: torch.nn.parameter.Parameter,
+        L_self_modules_features_modules_16_modules_branch1_modules_2_modules_bn_parameters_bias_: torch.nn.parameter.Parameter,
+        L_self_modules_features_modules_16_modules_branch2_modules_0_modules_conv_parameters_weight_: torch.nn.parameter.Parameter,
+        L_self_modules_features_modules_16_modules_branch2_modules_0_modules_bn_buffers_running_mean_: torch.Tensor,
+        L_self_modules_features_modules_16_modules_branch2_modules_0_modules_bn_buffers_running_var_: torch.Tensor,
+        L_self_modules_features_modules_16_modules_branch2_modules_0_modules_bn_parameters_weight_: torch.nn.parameter.Parameter,
+        L_self_modules_features_modules_16_modules_branch2_modules_0_modules_bn_parameters_bias_: torch.nn.parameter.Parameter,
+        L_self_modules_features_modules_16_modules_branch2_modules_1_modules_conv_parameters_weight_: torch.nn.parameter.Parameter,
+        L_self_modules_features_modules_16_modules_branch2_modules_1_modules_bn_buffers_running_mean_: torch.Tensor,
+        L_self_modules_features_modules_16_modules_branch2_modules_1_modules_bn_buffers_running_var_: torch.Tensor,
+        L_self_modules_features_modules_16_modules_branch2_modules_1_modules_bn_parameters_weight_: torch.nn.parameter.Parameter,
+        L_self_modules_features_modules_16_modules_branch2_modules_1_modules_bn_parameters_bias_: torch.nn.parameter.Parameter,
+        L_self_modules_features_modules_16_modules_branch2_modules_2_modules_conv_parameters_weight_: torch.nn.parameter.Parameter,
+        L_self_modules_features_modules_16_modules_branch2_modules_2_modules_bn_buffers_running_mean_: torch.Tensor,
+        L_self_modules_features_modules_16_modules_branch2_modules_2_modules_bn_buffers_running_var_: torch.Tensor,
+        L_self_modules_features_modules_16_modules_branch2_modules_2_modules_bn_parameters_weight_: torch.nn.parameter.Parameter,
+        L_self_modules_features_modules_16_modules_branch2_modules_2_modules_bn_parameters_bias_: torch.nn.parameter.Parameter,
+        L_self_modules_features_modules_16_modules_branch2_modules_3_modules_conv_parameters_weight_: torch.nn.parameter.Parameter,
+        L_self_modules_features_modules_16_modules_branch2_modules_3_modules_bn_buffers_running_mean_: torch.Tensor,
+        L_self_modules_features_modules_16_modules_branch2_modules_3_modules_bn_buffers_running_var_: torch.Tensor,
+        L_self_modules_features_modules_16_modules_branch2_modules_3_modules_bn_parameters_weight_: torch.nn.parameter.Parameter,
+        L_self_modules_features_modules_16_modules_branch2_modules_3_modules_bn_parameters_bias_: torch.nn.parameter.Parameter,
+        L_self_modules_features_modules_16_modules_branch2_modules_4_modules_conv_parameters_weight_: torch.nn.parameter.Parameter,
+        L_self_modules_features_modules_16_modules_branch2_modules_4_modules_bn_buffers_running_mean_: torch.Tensor,
+        L_self_modules_features_modules_16_modules_branch2_modules_4_modules_bn_buffers_running_var_: torch.Tensor,
+        L_self_modules_features_modules_16_modules_branch2_modules_4_modules_bn_parameters_weight_: torch.nn.parameter.Parameter,
+        L_self_modules_features_modules_16_modules_branch2_modules_4_modules_bn_parameters_bias_: torch.nn.parameter.Parameter,
+        L_self_modules_features_modules_16_modules_branch3_modules_1_modules_conv_parameters_weight_: torch.nn.parameter.Parameter,
+        L_self_modules_features_modules_16_modules_branch3_modules_1_modules_bn_buffers_running_mean_: torch.Tensor,
+        L_self_modules_features_modules_16_modules_branch3_modules_1_modules_bn_buffers_running_var_: torch.Tensor,
+        L_self_modules_features_modules_16_modules_branch3_modules_1_modules_bn_parameters_weight_: torch.nn.parameter.Parameter,
+        L_self_modules_features_modules_16_modules_branch3_modules_1_modules_bn_parameters_bias_: torch.nn.parameter.Parameter,
+        L_self_modules_features_modules_17_modules_branch0_modules_conv_parameters_weight_: torch.nn.parameter.Parameter,
+        L_self_modules_features_modules_17_modules_branch0_modules_bn_buffers_running_mean_: torch.Tensor,
+        L_self_modules_features_modules_17_modules_branch0_modules_bn_buffers_running_var_: torch.Tensor,
+        L_self_modules_features_modules_17_modules_branch0_modules_bn_parameters_weight_: torch.nn.parameter.Parameter,
+        L_self_modules_features_modules_17_modules_branch0_modules_bn_parameters_bias_: torch.nn.parameter.Parameter,
+        L_self_modules_features_modules_17_modules_branch1_modules_0_modules_conv_parameters_weight_: torch.nn.parameter.Parameter,
+        L_self_modules_features_modules_17_modules_branch1_modules_0_modules_bn_buffers_running_mean_: torch.Tensor,
+        L_self_modules_features_modules_17_modules_branch1_modules_0_modules_bn_buffers_running_var_: torch.Tensor,
+        L_self_modules_features_modules_17_modules_branch1_modules_0_modules_bn_parameters_weight_: torch.nn.parameter.Parameter,
+        L_self_modules_features_modules_17_modules_branch1_modules_0_modules_bn_parameters_bias_: torch.nn.parameter.Parameter,
+        L_self_modules_features_modules_17_modules_branch1_modules_1_modules_conv_parameters_weight_: torch.nn.parameter.Parameter,
+        L_self_modules_features_modules_17_modules_branch1_modules_1_modules_bn_buffers_running_mean_: torch.Tensor,
+        L_self_modules_features_modules_17_modules_branch1_modules_1_modules_bn_buffers_running_var_: torch.Tensor,
+        L_self_modules_features_modules_17_modules_branch1_modules_1_modules_bn_parameters_weight_: torch.nn.parameter.Parameter,
+        L_self_modules_features_modules_17_modules_branch1_modules_1_modules_bn_parameters_bias_: torch.nn.parameter.Parameter,
+        L_self_modules_features_modules_17_modules_branch1_modules_2_modules_conv_parameters_weight_: torch.nn.parameter.Parameter,
+        L_self_modules_features_modules_17_modules_branch1_modules_2_modules_bn_buffers_running_mean_: torch.Tensor,
+        L_self_modules_features_modules_17_modules_branch1_modules_2_modules_bn_buffers_running_var_: torch.Tensor,
+        L_self_modules_features_modules_17_modules_branch1_modules_2_modules_bn_parameters_weight_: torch.nn.parameter.Parameter,
+        L_self_modules_features_modules_17_modules_branch1_modules_2_modules_bn_parameters_bias_: torch.nn.parameter.Parameter,
+        L_self_modules_features_modules_17_modules_branch2_modules_0_modules_conv_parameters_weight_: torch.nn.parameter.Parameter,
+        L_self_modules_features_modules_17_modules_branch2_modules_0_modules_bn_buffers_running_mean_: torch.Tensor,
+        L_self_modules_features_modules_17_modules_branch2_modules_0_modules_bn_buffers_running_var_: torch.Tensor,
+        L_self_modules_features_modules_17_modules_branch2_modules_0_modules_bn_parameters_weight_: torch.nn.parameter.Parameter,
+        L_self_modules_features_modules_17_modules_branch2_modules_0_modules_bn_parameters_bias_: torch.nn.parameter.Parameter,
+        L_self_modules_features_modules_17_modules_branch2_modules_1_modules_conv_parameters_weight_: torch.nn.parameter.Parameter,
+        L_self_modules_features_modules_17_modules_branch2_modules_1_modules_bn_buffers_running_mean_: torch.Tensor,
+        L_self_modules_features_modules_17_modules_branch2_modules_1_modules_bn_buffers_running_var_: torch.Tensor,
+        L_self_modules_features_modules_17_modules_branch2_modules_1_modules_bn_parameters_weight_: torch.nn.parameter.Parameter,
+        L_self_modules_features_modules_17_modules_branch2_modules_1_modules_bn_parameters_bias_: torch.nn.parameter.Parameter,
+        L_self_modules_features_modules_17_modules_branch2_modules_2_modules_conv_parameters_weight_: torch.nn.parameter.Parameter,
+        L_self_modules_features_modules_17_modules_branch2_modules_2_modules_bn_buffers_running_mean_: torch.Tensor,
+        L_self_modules_features_modules_17_modules_branch2_modules_2_modules_bn_buffers_running_var_: torch.Tensor,
+        L_self_modules_features_modules_17_modules_branch2_modules_2_modules_bn_parameters_weight_: torch.nn.parameter.Parameter,
+        L_self_modules_features_modules_17_modules_branch2_modules_2_modules_bn_parameters_bias_: torch.nn.parameter.Parameter,
+        L_self_modules_features_modules_17_modules_branch2_modules_3_modules_conv_parameters_weight_: torch.nn.parameter.Parameter,
+        L_self_modules_features_modules_17_modules_branch2_modules_3_modules_bn_buffers_running_mean_: torch.Tensor,
+        L_self_modules_features_modules_17_modules_branch2_modules_3_modules_bn_buffers_running_var_: torch.Tensor,
+        L_self_modules_features_modules_17_modules_branch2_modules_3_modules_bn_parameters_weight_: torch.nn.parameter.Parameter,
+        L_self_modules_features_modules_17_modules_branch2_modules_3_modules_bn_parameters_bias_: torch.nn.parameter.Parameter,
+        L_self_modules_features_modules_17_modules_branch2_modules_4_modules_conv_parameters_weight_: torch.nn.parameter.Parameter,
+        L_self_modules_features_modules_17_modules_branch2_modules_4_modules_bn_buffers_running_mean_: torch.Tensor,
+        L_self_modules_features_modules_17_modules_branch2_modules_4_modules_bn_buffers_running_var_: torch.Tensor,
+        L_self_modules_features_modules_17_modules_branch2_modules_4_modules_bn_parameters_weight_: torch.nn.parameter.Parameter,
+        L_self_modules_features_modules_17_modules_branch2_modules_4_modules_bn_parameters_bias_: torch.nn.parameter.Parameter,
+        L_self_modules_features_modules_17_modules_branch3_modules_1_modules_conv_parameters_weight_: torch.nn.parameter.Parameter,
+        L_self_modules_features_modules_17_modules_branch3_modules_1_modules_bn_buffers_running_mean_: torch.Tensor,
+        L_self_modules_features_modules_17_modules_branch3_modules_1_modules_bn_buffers_running_var_: torch.Tensor,
+        L_self_modules_features_modules_17_modules_branch3_modules_1_modules_bn_parameters_weight_: torch.nn.parameter.Parameter,
+        L_self_modules_features_modules_17_modules_branch3_modules_1_modules_bn_parameters_bias_: torch.nn.parameter.Parameter,
+        L_self_modules_features_modules_18_modules_branch0_modules_0_modules_conv_parameters_weight_: torch.nn.parameter.Parameter,
+        L_self_modules_features_modules_18_modules_branch0_modules_0_modules_bn_buffers_running_mean_: torch.Tensor,
+        L_self_modules_features_modules_18_modules_branch0_modules_0_modules_bn_buffers_running_var_: torch.Tensor,
+        L_self_modules_features_modules_18_modules_branch0_modules_0_modules_bn_parameters_weight_: torch.nn.parameter.Parameter,
+        L_self_modules_features_modules_18_modules_branch0_modules_0_modules_bn_parameters_bias_: torch.nn.parameter.Parameter,
+        L_self_modules_features_modules_18_modules_branch0_modules_1_modules_conv_parameters_weight_: torch.nn.parameter.Parameter,
+        L_self_modules_features_modules_18_modules_branch0_modules_1_modules_bn_buffers_running_mean_: torch.Tensor,
+        L_self_modules_features_modules_18_modules_branch0_modules_1_modules_bn_buffers_running_var_: torch.Tensor,
+        L_self_modules_features_modules_18_modules_branch0_modules_1_modules_bn_parameters_weight_: torch.nn.parameter.Parameter,
+        L_self_modules_features_modules_18_modules_branch0_modules_1_modules_bn_parameters_bias_: torch.nn.parameter.Parameter,
+        L_self_modules_features_modules_18_modules_branch1_modules_0_modules_conv_parameters_weight_: torch.nn.parameter.Parameter,
+        L_self_modules_features_modules_18_modules_branch1_modules_0_modules_bn_buffers_running_mean_: torch.Tensor,
+        L_self_modules_features_modules_18_modules_branch1_modules_0_modules_bn_buffers_running_var_: torch.Tensor,
+        L_self_modules_features_modules_18_modules_branch1_modules_0_modules_bn_parameters_weight_: torch.nn.parameter.Parameter,
+        L_self_modules_features_modules_18_modules_branch1_modules_0_modules_bn_parameters_bias_: torch.nn.parameter.Parameter,
+        L_self_modules_features_modules_18_modules_branch1_modules_1_modules_conv_parameters_weight_: torch.nn.parameter.Parameter,
+        L_self_modules_features_modules_18_modules_branch1_modules_1_modules_bn_buffers_running_mean_: torch.Tensor,
+        L_self_modules_features_modules_18_modules_branch1_modules_1_modules_bn_buffers_running_var_: torch.Tensor,
+        L_self_modules_features_modules_18_modules_branch1_modules_1_modules_bn_parameters_weight_: torch.nn.parameter.Parameter,
+        L_self_modules_features_modules_18_modules_branch1_modules_1_modules_bn_parameters_bias_: torch.nn.parameter.Parameter,
+        L_self_modules_features_modules_18_modules_branch1_modules_2_modules_conv_parameters_weight_: torch.nn.parameter.Parameter,
+        L_self_modules_features_modules_18_modules_branch1_modules_2_modules_bn_buffers_running_mean_: torch.Tensor,
+        L_self_modules_features_modules_18_modules_branch1_modules_2_modules_bn_buffers_running_var_: torch.Tensor,
+        L_self_modules_features_modules_18_modules_branch1_modules_2_modules_bn_parameters_weight_: torch.nn.parameter.Parameter,
+        L_self_modules_features_modules_18_modules_branch1_modules_2_modules_bn_parameters_bias_: torch.nn.parameter.Parameter,
+        L_self_modules_features_modules_18_modules_branch1_modules_3_modules_conv_parameters_weight_: torch.nn.parameter.Parameter,
+        L_self_modules_features_modules_18_modules_branch1_modules_3_modules_bn_buffers_running_mean_: torch.Tensor,
+        L_self_modules_features_modules_18_modules_branch1_modules_3_modules_bn_buffers_running_var_: torch.Tensor,
+        L_self_modules_features_modules_18_modules_branch1_modules_3_modules_bn_parameters_weight_: torch.nn.parameter.Parameter,
+        L_self_modules_features_modules_18_modules_branch1_modules_3_modules_bn_parameters_bias_: torch.nn.parameter.Parameter,
+        L_self_modules_features_modules_19_modules_branch0_modules_conv_parameters_weight_: torch.nn.parameter.Parameter,
+        L_self_modules_features_modules_19_modules_branch0_modules_bn_buffers_running_mean_: torch.Tensor,
+        L_self_modules_features_modules_19_modules_branch0_modules_bn_buffers_running_var_: torch.Tensor,
+        L_self_modules_features_modules_19_modules_branch0_modules_bn_parameters_weight_: torch.nn.parameter.Parameter,
+        L_self_modules_features_modules_19_modules_branch0_modules_bn_parameters_bias_: torch.nn.parameter.Parameter,
+        L_self_modules_features_modules_19_modules_branch1_0_modules_conv_parameters_weight_: torch.nn.parameter.Parameter,
+        L_self_modules_features_modules_19_modules_branch1_0_modules_bn_buffers_running_mean_: torch.Tensor,
+        L_self_modules_features_modules_19_modules_branch1_0_modules_bn_buffers_running_var_: torch.Tensor,
+        L_self_modules_features_modules_19_modules_branch1_0_modules_bn_parameters_weight_: torch.nn.parameter.Parameter,
+        L_self_modules_features_modules_19_modules_branch1_0_modules_bn_parameters_bias_: torch.nn.parameter.Parameter,
+        L_self_modules_features_modules_19_modules_branch1_1a_modules_conv_parameters_weight_: torch.nn.parameter.Parameter,
+        L_self_modules_features_modules_19_modules_branch1_1a_modules_bn_buffers_running_mean_: torch.Tensor,
+        L_self_modules_features_modules_19_modules_branch1_1a_modules_bn_buffers_running_var_: torch.Tensor,
+        L_self_modules_features_modules_19_modules_branch1_1a_modules_bn_parameters_weight_: torch.nn.parameter.Parameter,
+        L_self_modules_features_modules_19_modules_branch1_1a_modules_bn_parameters_bias_: torch.nn.parameter.Parameter,
+        L_self_modules_features_modules_19_modules_branch1_1b_modules_conv_parameters_weight_: torch.nn.parameter.Parameter,
+        L_self_modules_features_modules_19_modules_branch1_1b_modules_bn_buffers_running_mean_: torch.Tensor,
+        L_self_modules_features_modules_19_modules_branch1_1b_modules_bn_buffers_running_var_: torch.Tensor,
+        L_self_modules_features_modules_19_modules_branch1_1b_modules_bn_parameters_weight_: torch.nn.parameter.Parameter,
+        L_self_modules_features_modules_19_modules_branch1_1b_modules_bn_parameters_bias_: torch.nn.parameter.Parameter,
+        L_self_modules_features_modules_19_modules_branch2_0_modules_conv_parameters_weight_: torch.nn.parameter.Parameter,
+        L_self_modules_features_modules_19_modules_branch2_0_modules_bn_buffers_running_mean_: torch.Tensor,
+        L_self_modules_features_modules_19_modules_branch2_0_modules_bn_buffers_running_var_: torch.Tensor,
+        L_self_modules_features_modules_19_modules_branch2_0_modules_bn_parameters_weight_: torch.nn.parameter.Parameter,
+        L_self_modules_features_modules_19_modules_branch2_0_modules_bn_parameters_bias_: torch.nn.parameter.Parameter,
+        L_self_modules_features_modules_19_modules_branch2_1_modules_conv_parameters_weight_: torch.nn.parameter.Parameter,
+        L_self_modules_features_modules_19_modules_branch2_1_modules_bn_buffers_running_mean_: torch.Tensor,
+        L_self_modules_features_modules_19_modules_branch2_1_modules_bn_buffers_running_var_: torch.Tensor,
+        L_self_modules_features_modules_19_modules_branch2_1_modules_bn_parameters_weight_: torch.nn.parameter.Parameter,
+        L_self_modules_features_modules_19_modules_branch2_1_modules_bn_parameters_bias_: torch.nn.parameter.Parameter,
+        L_self_modules_features_modules_19_modules_branch2_2_modules_conv_parameters_weight_: torch.nn.parameter.Parameter,
+        L_self_modules_features_modules_19_modules_branch2_2_modules_bn_buffers_running_mean_: torch.Tensor,
+        L_self_modules_features_modules_19_modules_branch2_2_modules_bn_buffers_running_var_: torch.Tensor,
+        L_self_modules_features_modules_19_modules_branch2_2_modules_bn_parameters_weight_: torch.nn.parameter.Parameter,
+        L_self_modules_features_modules_19_modules_branch2_2_modules_bn_parameters_bias_: torch.nn.parameter.Parameter,
+        L_self_modules_features_modules_19_modules_branch2_3a_modules_conv_parameters_weight_: torch.nn.parameter.Parameter,
+        L_self_modules_features_modules_19_modules_branch2_3a_modules_bn_buffers_running_mean_: torch.Tensor,
+        L_self_modules_features_modules_19_modules_branch2_3a_modules_bn_buffers_running_var_: torch.Tensor,
+        L_self_modules_features_modules_19_modules_branch2_3a_modules_bn_parameters_weight_: torch.nn.parameter.Parameter,
+        L_self_modules_features_modules_19_modules_branch2_3a_modules_bn_parameters_bias_: torch.nn.parameter.Parameter,
+        L_self_modules_features_modules_19_modules_branch2_3b_modules_conv_parameters_weight_: torch.nn.parameter.Parameter,
+        L_self_modules_features_modules_19_modules_branch2_3b_modules_bn_buffers_running_mean_: torch.Tensor,
+        L_self_modules_features_modules_19_modules_branch2_3b_modules_bn_buffers_running_var_: torch.Tensor,
+        L_self_modules_features_modules_19_modules_branch2_3b_modules_bn_parameters_weight_: torch.nn.parameter.Parameter,
+        L_self_modules_features_modules_19_modules_branch2_3b_modules_bn_parameters_bias_: torch.nn.parameter.Parameter,
+        L_self_modules_features_modules_19_modules_branch3_modules_1_modules_conv_parameters_weight_: torch.nn.parameter.Parameter,
+        L_self_modules_features_modules_19_modules_branch3_modules_1_modules_bn_buffers_running_mean_: torch.Tensor,
+        L_self_modules_features_modules_19_modules_branch3_modules_1_modules_bn_buffers_running_var_: torch.Tensor,
+        L_self_modules_features_modules_19_modules_branch3_modules_1_modules_bn_parameters_weight_: torch.nn.parameter.Parameter,
+        L_self_modules_features_modules_19_modules_branch3_modules_1_modules_bn_parameters_bias_: torch.nn.parameter.Parameter,
+        L_self_modules_features_modules_20_modules_branch0_modules_conv_parameters_weight_: torch.nn.parameter.Parameter,
+        L_self_modules_features_modules_20_modules_branch0_modules_bn_buffers_running_mean_: torch.Tensor,
+        L_self_modules_features_modules_20_modules_branch0_modules_bn_buffers_running_var_: torch.Tensor,
+        L_self_modules_features_modules_20_modules_branch0_modules_bn_parameters_weight_: torch.nn.parameter.Parameter,
+        L_self_modules_features_modules_20_modules_branch0_modules_bn_parameters_bias_: torch.nn.parameter.Parameter,
+        L_self_modules_features_modules_20_modules_branch1_0_modules_conv_parameters_weight_: torch.nn.parameter.Parameter,
+        L_self_modules_features_modules_20_modules_branch1_0_modules_bn_buffers_running_mean_: torch.Tensor,
+        L_self_modules_features_modules_20_modules_branch1_0_modules_bn_buffers_running_var_: torch.Tensor,
+        L_self_modules_features_modules_20_modules_branch1_0_modules_bn_parameters_weight_: torch.nn.parameter.Parameter,
+        L_self_modules_features_modules_20_modules_branch1_0_modules_bn_parameters_bias_: torch.nn.parameter.Parameter,
+        L_self_modules_features_modules_20_modules_branch1_1a_modules_conv_parameters_weight_: torch.nn.parameter.Parameter,
+        L_self_modules_features_modules_20_modules_branch1_1a_modules_bn_buffers_running_mean_: torch.Tensor,
+        L_self_modules_features_modules_20_modules_branch1_1a_modules_bn_buffers_running_var_: torch.Tensor,
+        L_self_modules_features_modules_20_modules_branch1_1a_modules_bn_parameters_weight_: torch.nn.parameter.Parameter,
+        L_self_modules_features_modules_20_modules_branch1_1a_modules_bn_parameters_bias_: torch.nn.parameter.Parameter,
+        L_self_modules_features_modules_20_modules_branch1_1b_modules_conv_parameters_weight_: torch.nn.parameter.Parameter,
+        L_self_modules_features_modules_20_modules_branch1_1b_modules_bn_buffers_running_mean_: torch.Tensor,
+        L_self_modules_features_modules_20_modules_branch1_1b_modules_bn_buffers_running_var_: torch.Tensor,
+        L_self_modules_features_modules_20_modules_branch1_1b_modules_bn_parameters_weight_: torch.nn.parameter.Parameter,
+        L_self_modules_features_modules_20_modules_branch1_1b_modules_bn_parameters_bias_: torch.nn.parameter.Parameter,
+        L_self_modules_features_modules_20_modules_branch2_0_modules_conv_parameters_weight_: torch.nn.parameter.Parameter,
+        L_self_modules_features_modules_20_modules_branch2_0_modules_bn_buffers_running_mean_: torch.Tensor,
+        L_self_modules_features_modules_20_modules_branch2_0_modules_bn_buffers_running_var_: torch.Tensor,
+        L_self_modules_features_modules_20_modules_branch2_0_modules_bn_parameters_weight_: torch.nn.parameter.Parameter,
+        L_self_modules_features_modules_20_modules_branch2_0_modules_bn_parameters_bias_: torch.nn.parameter.Parameter,
+        L_self_modules_features_modules_20_modules_branch2_1_modules_conv_parameters_weight_: torch.nn.parameter.Parameter,
+        L_self_modules_features_modules_20_modules_branch2_1_modules_bn_buffers_running_mean_: torch.Tensor,
+        L_self_modules_features_modules_20_modules_branch2_1_modules_bn_buffers_running_var_: torch.Tensor,
+        L_self_modules_features_modules_20_modules_branch2_1_modules_bn_parameters_weight_: torch.nn.parameter.Parameter,
+        L_self_modules_features_modules_20_modules_branch2_1_modules_bn_parameters_bias_: torch.nn.parameter.Parameter,
+        L_self_modules_features_modules_20_modules_branch2_2_modules_conv_parameters_weight_: torch.nn.parameter.Parameter,
+        L_self_modules_features_modules_20_modules_branch2_2_modules_bn_buffers_running_mean_: torch.Tensor,
+        L_self_modules_features_modules_20_modules_branch2_2_modules_bn_buffers_running_var_: torch.Tensor,
+        L_self_modules_features_modules_20_modules_branch2_2_modules_bn_parameters_weight_: torch.nn.parameter.Parameter,
+        L_self_modules_features_modules_20_modules_branch2_2_modules_bn_parameters_bias_: torch.nn.parameter.Parameter,
+        L_self_modules_features_modules_20_modules_branch2_3a_modules_conv_parameters_weight_: torch.nn.parameter.Parameter,
+        L_self_modules_features_modules_20_modules_branch2_3a_modules_bn_buffers_running_mean_: torch.Tensor,
+        L_self_modules_features_modules_20_modules_branch2_3a_modules_bn_buffers_running_var_: torch.Tensor,
+        L_self_modules_features_modules_20_modules_branch2_3a_modules_bn_parameters_weight_: torch.nn.parameter.Parameter,
+        L_self_modules_features_modules_20_modules_branch2_3a_modules_bn_parameters_bias_: torch.nn.parameter.Parameter,
+        L_self_modules_features_modules_20_modules_branch2_3b_modules_conv_parameters_weight_: torch.nn.parameter.Parameter,
+        L_self_modules_features_modules_20_modules_branch2_3b_modules_bn_buffers_running_mean_: torch.Tensor,
+        L_self_modules_features_modules_20_modules_branch2_3b_modules_bn_buffers_running_var_: torch.Tensor,
+        L_self_modules_features_modules_20_modules_branch2_3b_modules_bn_parameters_weight_: torch.nn.parameter.Parameter,
+        L_self_modules_features_modules_20_modules_branch2_3b_modules_bn_parameters_bias_: torch.nn.parameter.Parameter,
+        L_self_modules_features_modules_20_modules_branch3_modules_1_modules_conv_parameters_weight_: torch.nn.parameter.Parameter,
+        L_self_modules_features_modules_20_modules_branch3_modules_1_modules_bn_buffers_running_mean_: torch.Tensor,
+        L_self_modules_features_modules_20_modules_branch3_modules_1_modules_bn_buffers_running_var_: torch.Tensor,
+        L_self_modules_features_modules_20_modules_branch3_modules_1_modules_bn_parameters_weight_: torch.nn.parameter.Parameter,
+        L_self_modules_features_modules_20_modules_branch3_modules_1_modules_bn_parameters_bias_: torch.nn.parameter.Parameter,
+        L_self_modules_features_modules_21_modules_branch0_modules_conv_parameters_weight_: torch.nn.parameter.Parameter,
+        L_self_modules_features_modules_21_modules_branch0_modules_bn_buffers_running_mean_: torch.Tensor,
+        L_self_modules_features_modules_21_modules_branch0_modules_bn_buffers_running_var_: torch.Tensor,
+        L_self_modules_features_modules_21_modules_branch0_modules_bn_parameters_weight_: torch.nn.parameter.Parameter,
+        L_self_modules_features_modules_21_modules_branch0_modules_bn_parameters_bias_: torch.nn.parameter.Parameter,
+        L_self_modules_features_modules_21_modules_branch1_0_modules_conv_parameters_weight_: torch.nn.parameter.Parameter,
+        L_self_modules_features_modules_21_modules_branch1_0_modules_bn_buffers_running_mean_: torch.Tensor,
+        L_self_modules_features_modules_21_modules_branch1_0_modules_bn_buffers_running_var_: torch.Tensor,
+        L_self_modules_features_modules_21_modules_branch1_0_modules_bn_parameters_weight_: torch.nn.parameter.Parameter,
+        L_self_modules_features_modules_21_modules_branch1_0_modules_bn_parameters_bias_: torch.nn.parameter.Parameter,
+        L_self_modules_features_modules_21_modules_branch1_1a_modules_conv_parameters_weight_: torch.nn.parameter.Parameter,
+        L_self_modules_features_modules_21_modules_branch1_1a_modules_bn_buffers_running_mean_: torch.Tensor,
+        L_self_modules_features_modules_21_modules_branch1_1a_modules_bn_buffers_running_var_: torch.Tensor,
+        L_self_modules_features_modules_21_modules_branch1_1a_modules_bn_parameters_weight_: torch.nn.parameter.Parameter,
+        L_self_modules_features_modules_21_modules_branch1_1a_modules_bn_parameters_bias_: torch.nn.parameter.Parameter,
+        L_self_modules_features_modules_21_modules_branch1_1b_modules_conv_parameters_weight_: torch.nn.parameter.Parameter,
+        L_self_modules_features_modules_21_modules_branch1_1b_modules_bn_buffers_running_mean_: torch.Tensor,
+        L_self_modules_features_modules_21_modules_branch1_1b_modules_bn_buffers_running_var_: torch.Tensor,
+        L_self_modules_features_modules_21_modules_branch1_1b_modules_bn_parameters_weight_: torch.nn.parameter.Parameter,
+        L_self_modules_features_modules_21_modules_branch1_1b_modules_bn_parameters_bias_: torch.nn.parameter.Parameter,
+        L_self_modules_features_modules_21_modules_branch2_0_modules_conv_parameters_weight_: torch.nn.parameter.Parameter,
+        L_self_modules_features_modules_21_modules_branch2_0_modules_bn_buffers_running_mean_: torch.Tensor,
+        L_self_modules_features_modules_21_modules_branch2_0_modules_bn_buffers_running_var_: torch.Tensor,
+        L_self_modules_features_modules_21_modules_branch2_0_modules_bn_parameters_weight_: torch.nn.parameter.Parameter,
+        L_self_modules_features_modules_21_modules_branch2_0_modules_bn_parameters_bias_: torch.nn.parameter.Parameter,
+        L_self_modules_features_modules_21_modules_branch2_1_modules_conv_parameters_weight_: torch.nn.parameter.Parameter,
+        L_self_modules_features_modules_21_modules_branch2_1_modules_bn_buffers_running_mean_: torch.Tensor,
+        L_self_modules_features_modules_21_modules_branch2_1_modules_bn_buffers_running_var_: torch.Tensor,
+        L_self_modules_features_modules_21_modules_branch2_1_modules_bn_parameters_weight_: torch.nn.parameter.Parameter,
+        L_self_modules_features_modules_21_modules_branch2_1_modules_bn_parameters_bias_: torch.nn.parameter.Parameter,
+        L_self_modules_features_modules_21_modules_branch2_2_modules_conv_parameters_weight_: torch.nn.parameter.Parameter,
+        L_self_modules_features_modules_21_modules_branch2_2_modules_bn_buffers_running_mean_: torch.Tensor,
+        L_self_modules_features_modules_21_modules_branch2_2_modules_bn_buffers_running_var_: torch.Tensor,
+        L_self_modules_features_modules_21_modules_branch2_2_modules_bn_parameters_weight_: torch.nn.parameter.Parameter,
+        L_self_modules_features_modules_21_modules_branch2_2_modules_bn_parameters_bias_: torch.nn.parameter.Parameter,
+        L_self_modules_features_modules_21_modules_branch2_3a_modules_conv_parameters_weight_: torch.nn.parameter.Parameter,
+        L_self_modules_features_modules_21_modules_branch2_3a_modules_bn_buffers_running_mean_: torch.Tensor,
+        L_self_modules_features_modules_21_modules_branch2_3a_modules_bn_buffers_running_var_: torch.Tensor,
+        L_self_modules_features_modules_21_modules_branch2_3a_modules_bn_parameters_weight_: torch.nn.parameter.Parameter,
+        L_self_modules_features_modules_21_modules_branch2_3a_modules_bn_parameters_bias_: torch.nn.parameter.Parameter,
+        L_self_modules_features_modules_21_modules_branch2_3b_modules_conv_parameters_weight_: torch.nn.parameter.Parameter,
+        L_self_modules_features_modules_21_modules_branch2_3b_modules_bn_buffers_running_mean_: torch.Tensor,
+        L_self_modules_features_modules_21_modules_branch2_3b_modules_bn_buffers_running_var_: torch.Tensor,
+        L_self_modules_features_modules_21_modules_branch2_3b_modules_bn_parameters_weight_: torch.nn.parameter.Parameter,
+        L_self_modules_features_modules_21_modules_branch2_3b_modules_bn_parameters_bias_: torch.nn.parameter.Parameter,
+        L_self_modules_features_modules_21_modules_branch3_modules_1_modules_conv_parameters_weight_: torch.nn.parameter.Parameter,
+        L_self_modules_features_modules_21_modules_branch3_modules_1_modules_bn_buffers_running_mean_: torch.Tensor,
+        L_self_modules_features_modules_21_modules_branch3_modules_1_modules_bn_buffers_running_var_: torch.Tensor,
+        L_self_modules_features_modules_21_modules_branch3_modules_1_modules_bn_parameters_weight_: torch.nn.parameter.Parameter,
+        L_self_modules_features_modules_21_modules_branch3_modules_1_modules_bn_parameters_bias_: torch.nn.parameter.Parameter,
+        L_self_modules_last_linear_parameters_weight_: torch.nn.parameter.Parameter,
+        L_self_modules_last_linear_parameters_bias_: torch.nn.parameter.Parameter,
+    ):
+        l_self_modules_features_modules_0_modules_conv_parameters_weight_ = (
+            L_self_modules_features_modules_0_modules_conv_parameters_weight_
+        )
+        l_x_ = L_x_
+        l_self_modules_features_modules_0_modules_bn_buffers_running_mean_ = (
+            L_self_modules_features_modules_0_modules_bn_buffers_running_mean_
+        )
+        l_self_modules_features_modules_0_modules_bn_buffers_running_var_ = (
+            L_self_modules_features_modules_0_modules_bn_buffers_running_var_
+        )
+        l_self_modules_features_modules_0_modules_bn_parameters_weight_ = (
+            L_self_modules_features_modules_0_modules_bn_parameters_weight_
+        )
+        l_self_modules_features_modules_0_modules_bn_parameters_bias_ = (
+            L_self_modules_features_modules_0_modules_bn_parameters_bias_
+        )
+        l_self_modules_features_modules_1_modules_conv_parameters_weight_ = (
+            L_self_modules_features_modules_1_modules_conv_parameters_weight_
+        )
+        l_self_modules_features_modules_1_modules_bn_buffers_running_mean_ = (
+            L_self_modules_features_modules_1_modules_bn_buffers_running_mean_
+        )
+        l_self_modules_features_modules_1_modules_bn_buffers_running_var_ = (
+            L_self_modules_features_modules_1_modules_bn_buffers_running_var_
+        )
+        l_self_modules_features_modules_1_modules_bn_parameters_weight_ = (
+            L_self_modules_features_modules_1_modules_bn_parameters_weight_
+        )
+        l_self_modules_features_modules_1_modules_bn_parameters_bias_ = (
+            L_self_modules_features_modules_1_modules_bn_parameters_bias_
+        )
+        l_self_modules_features_modules_2_modules_conv_parameters_weight_ = (
+            L_self_modules_features_modules_2_modules_conv_parameters_weight_
+        )
+        l_self_modules_features_modules_2_modules_bn_buffers_running_mean_ = (
+            L_self_modules_features_modules_2_modules_bn_buffers_running_mean_
+        )
+        l_self_modules_features_modules_2_modules_bn_buffers_running_var_ = (
+            L_self_modules_features_modules_2_modules_bn_buffers_running_var_
+        )
+        l_self_modules_features_modules_2_modules_bn_parameters_weight_ = (
+            L_self_modules_features_modules_2_modules_bn_parameters_weight_
+        )
+        l_self_modules_features_modules_2_modules_bn_parameters_bias_ = (
+            L_self_modules_features_modules_2_modules_bn_parameters_bias_
+        )
+        l_self_modules_features_modules_3_modules_conv_modules_conv_parameters_weight_ = L_self_modules_features_modules_3_modules_conv_modules_conv_parameters_weight_
+        l_self_modules_features_modules_3_modules_conv_modules_bn_buffers_running_mean_ = L_self_modules_features_modules_3_modules_conv_modules_bn_buffers_running_mean_
+        l_self_modules_features_modules_3_modules_conv_modules_bn_buffers_running_var_ = L_self_modules_features_modules_3_modules_conv_modules_bn_buffers_running_var_
+        l_self_modules_features_modules_3_modules_conv_modules_bn_parameters_weight_ = (
+            L_self_modules_features_modules_3_modules_conv_modules_bn_parameters_weight_
+        )
+        l_self_modules_features_modules_3_modules_conv_modules_bn_parameters_bias_ = (
+            L_self_modules_features_modules_3_modules_conv_modules_bn_parameters_bias_
+        )
+        l_self_modules_features_modules_4_modules_branch0_modules_0_modules_conv_parameters_weight_ = L_self_modules_features_modules_4_modules_branch0_modules_0_modules_conv_parameters_weight_
+        l_self_modules_features_modules_4_modules_branch0_modules_0_modules_bn_buffers_running_mean_ = L_self_modules_features_modules_4_modules_branch0_modules_0_modules_bn_buffers_running_mean_
+        l_self_modules_features_modules_4_modules_branch0_modules_0_modules_bn_buffers_running_var_ = L_self_modules_features_modules_4_modules_branch0_modules_0_modules_bn_buffers_running_var_
+        l_self_modules_features_modules_4_modules_branch0_modules_0_modules_bn_parameters_weight_ = L_self_modules_features_modules_4_modules_branch0_modules_0_modules_bn_parameters_weight_
+        l_self_modules_features_modules_4_modules_branch0_modules_0_modules_bn_parameters_bias_ = L_self_modules_features_modules_4_modules_branch0_modules_0_modules_bn_parameters_bias_
+        l_self_modules_features_modules_4_modules_branch0_modules_1_modules_conv_parameters_weight_ = L_self_modules_features_modules_4_modules_branch0_modules_1_modules_conv_parameters_weight_
+        l_self_modules_features_modules_4_modules_branch0_modules_1_modules_bn_buffers_running_mean_ = L_self_modules_features_modules_4_modules_branch0_modules_1_modules_bn_buffers_running_mean_
+        l_self_modules_features_modules_4_modules_branch0_modules_1_modules_bn_buffers_running_var_ = L_self_modules_features_modules_4_modules_branch0_modules_1_modules_bn_buffers_running_var_
+        l_self_modules_features_modules_4_modules_branch0_modules_1_modules_bn_parameters_weight_ = L_self_modules_features_modules_4_modules_branch0_modules_1_modules_bn_parameters_weight_
+        l_self_modules_features_modules_4_modules_branch0_modules_1_modules_bn_parameters_bias_ = L_self_modules_features_modules_4_modules_branch0_modules_1_modules_bn_parameters_bias_
+        l_self_modules_features_modules_4_modules_branch1_modules_0_modules_conv_parameters_weight_ = L_self_modules_features_modules_4_modules_branch1_modules_0_modules_conv_parameters_weight_
+        l_self_modules_features_modules_4_modules_branch1_modules_0_modules_bn_buffers_running_mean_ = L_self_modules_features_modules_4_modules_branch1_modules_0_modules_bn_buffers_running_mean_
+        l_self_modules_features_modules_4_modules_branch1_modules_0_modules_bn_buffers_running_var_ = L_self_modules_features_modules_4_modules_branch1_modules_0_modules_bn_buffers_running_var_
+        l_self_modules_features_modules_4_modules_branch1_modules_0_modules_bn_parameters_weight_ = L_self_modules_features_modules_4_modules_branch1_modules_0_modules_bn_parameters_weight_
+        l_self_modules_features_modules_4_modules_branch1_modules_0_modules_bn_parameters_bias_ = L_self_modules_features_modules_4_modules_branch1_modules_0_modules_bn_parameters_bias_
+        l_self_modules_features_modules_4_modules_branch1_modules_1_modules_conv_parameters_weight_ = L_self_modules_features_modules_4_modules_branch1_modules_1_modules_conv_parameters_weight_
+        l_self_modules_features_modules_4_modules_branch1_modules_1_modules_bn_buffers_running_mean_ = L_self_modules_features_modules_4_modules_branch1_modules_1_modules_bn_buffers_running_mean_
+        l_self_modules_features_modules_4_modules_branch1_modules_1_modules_bn_buffers_running_var_ = L_self_modules_features_modules_4_modules_branch1_modules_1_modules_bn_buffers_running_var_
+        l_self_modules_features_modules_4_modules_branch1_modules_1_modules_bn_parameters_weight_ = L_self_modules_features_modules_4_modules_branch1_modules_1_modules_bn_parameters_weight_
+        l_self_modules_features_modules_4_modules_branch1_modules_1_modules_bn_parameters_bias_ = L_self_modules_features_modules_4_modules_branch1_modules_1_modules_bn_parameters_bias_
+        l_self_modules_features_modules_4_modules_branch1_modules_2_modules_conv_parameters_weight_ = L_self_modules_features_modules_4_modules_branch1_modules_2_modules_conv_parameters_weight_
+        l_self_modules_features_modules_4_modules_branch1_modules_2_modules_bn_buffers_running_mean_ = L_self_modules_features_modules_4_modules_branch1_modules_2_modules_bn_buffers_running_mean_
+        l_self_modules_features_modules_4_modules_branch1_modules_2_modules_bn_buffers_running_var_ = L_self_modules_features_modules_4_modules_branch1_modules_2_modules_bn_buffers_running_var_
+        l_self_modules_features_modules_4_modules_branch1_modules_2_modules_bn_parameters_weight_ = L_self_modules_features_modules_4_modules_branch1_modules_2_modules_bn_parameters_weight_
+        l_self_modules_features_modules_4_modules_branch1_modules_2_modules_bn_parameters_bias_ = L_self_modules_features_modules_4_modules_branch1_modules_2_modules_bn_parameters_bias_
+        l_self_modules_features_modules_4_modules_branch1_modules_3_modules_conv_parameters_weight_ = L_self_modules_features_modules_4_modules_branch1_modules_3_modules_conv_parameters_weight_
+        l_self_modules_features_modules_4_modules_branch1_modules_3_modules_bn_buffers_running_mean_ = L_self_modules_features_modules_4_modules_branch1_modules_3_modules_bn_buffers_running_mean_
+        l_self_modules_features_modules_4_modules_branch1_modules_3_modules_bn_buffers_running_var_ = L_self_modules_features_modules_4_modules_branch1_modules_3_modules_bn_buffers_running_var_
+        l_self_modules_features_modules_4_modules_branch1_modules_3_modules_bn_parameters_weight_ = L_self_modules_features_modules_4_modules_branch1_modules_3_modules_bn_parameters_weight_
+        l_self_modules_features_modules_4_modules_branch1_modules_3_modules_bn_parameters_bias_ = L_self_modules_features_modules_4_modules_branch1_modules_3_modules_bn_parameters_bias_
+        l_self_modules_features_modules_5_modules_conv_modules_conv_parameters_weight_ = L_self_modules_features_modules_5_modules_conv_modules_conv_parameters_weight_
+        l_self_modules_features_modules_5_modules_conv_modules_bn_buffers_running_mean_ = L_self_modules_features_modules_5_modules_conv_modules_bn_buffers_running_mean_
+        l_self_modules_features_modules_5_modules_conv_modules_bn_buffers_running_var_ = L_self_modules_features_modules_5_modules_conv_modules_bn_buffers_running_var_
+        l_self_modules_features_modules_5_modules_conv_modules_bn_parameters_weight_ = (
+            L_self_modules_features_modules_5_modules_conv_modules_bn_parameters_weight_
+        )
+        l_self_modules_features_modules_5_modules_conv_modules_bn_parameters_bias_ = (
+            L_self_modules_features_modules_5_modules_conv_modules_bn_parameters_bias_
+        )
+        l_self_modules_features_modules_6_modules_branch0_modules_conv_parameters_weight_ = L_self_modules_features_modules_6_modules_branch0_modules_conv_parameters_weight_
+        l_self_modules_features_modules_6_modules_branch0_modules_bn_buffers_running_mean_ = L_self_modules_features_modules_6_modules_branch0_modules_bn_buffers_running_mean_
+        l_self_modules_features_modules_6_modules_branch0_modules_bn_buffers_running_var_ = L_self_modules_features_modules_6_modules_branch0_modules_bn_buffers_running_var_
+        l_self_modules_features_modules_6_modules_branch0_modules_bn_parameters_weight_ = L_self_modules_features_modules_6_modules_branch0_modules_bn_parameters_weight_
+        l_self_modules_features_modules_6_modules_branch0_modules_bn_parameters_bias_ = L_self_modules_features_modules_6_modules_branch0_modules_bn_parameters_bias_
+        l_self_modules_features_modules_6_modules_branch1_modules_0_modules_conv_parameters_weight_ = L_self_modules_features_modules_6_modules_branch1_modules_0_modules_conv_parameters_weight_
+        l_self_modules_features_modules_6_modules_branch1_modules_0_modules_bn_buffers_running_mean_ = L_self_modules_features_modules_6_modules_branch1_modules_0_modules_bn_buffers_running_mean_
+        l_self_modules_features_modules_6_modules_branch1_modules_0_modules_bn_buffers_running_var_ = L_self_modules_features_modules_6_modules_branch1_modules_0_modules_bn_buffers_running_var_
+        l_self_modules_features_modules_6_modules_branch1_modules_0_modules_bn_parameters_weight_ = L_self_modules_features_modules_6_modules_branch1_modules_0_modules_bn_parameters_weight_
+        l_self_modules_features_modules_6_modules_branch1_modules_0_modules_bn_parameters_bias_ = L_self_modules_features_modules_6_modules_branch1_modules_0_modules_bn_parameters_bias_
+        l_self_modules_features_modules_6_modules_branch1_modules_1_modules_conv_parameters_weight_ = L_self_modules_features_modules_6_modules_branch1_modules_1_modules_conv_parameters_weight_
+        l_self_modules_features_modules_6_modules_branch1_modules_1_modules_bn_buffers_running_mean_ = L_self_modules_features_modules_6_modules_branch1_modules_1_modules_bn_buffers_running_mean_
+        l_self_modules_features_modules_6_modules_branch1_modules_1_modules_bn_buffers_running_var_ = L_self_modules_features_modules_6_modules_branch1_modules_1_modules_bn_buffers_running_var_
+        l_self_modules_features_modules_6_modules_branch1_modules_1_modules_bn_parameters_weight_ = L_self_modules_features_modules_6_modules_branch1_modules_1_modules_bn_parameters_weight_
+        l_self_modules_features_modules_6_modules_branch1_modules_1_modules_bn_parameters_bias_ = L_self_modules_features_modules_6_modules_branch1_modules_1_modules_bn_parameters_bias_
+        l_self_modules_features_modules_6_modules_branch2_modules_0_modules_conv_parameters_weight_ = L_self_modules_features_modules_6_modules_branch2_modules_0_modules_conv_parameters_weight_
+        l_self_modules_features_modules_6_modules_branch2_modules_0_modules_bn_buffers_running_mean_ = L_self_modules_features_modules_6_modules_branch2_modules_0_modules_bn_buffers_running_mean_
+        l_self_modules_features_modules_6_modules_branch2_modules_0_modules_bn_buffers_running_var_ = L_self_modules_features_modules_6_modules_branch2_modules_0_modules_bn_buffers_running_var_
+        l_self_modules_features_modules_6_modules_branch2_modules_0_modules_bn_parameters_weight_ = L_self_modules_features_modules_6_modules_branch2_modules_0_modules_bn_parameters_weight_
+        l_self_modules_features_modules_6_modules_branch2_modules_0_modules_bn_parameters_bias_ = L_self_modules_features_modules_6_modules_branch2_modules_0_modules_bn_parameters_bias_
+        l_self_modules_features_modules_6_modules_branch2_modules_1_modules_conv_parameters_weight_ = L_self_modules_features_modules_6_modules_branch2_modules_1_modules_conv_parameters_weight_
+        l_self_modules_features_modules_6_modules_branch2_modules_1_modules_bn_buffers_running_mean_ = L_self_modules_features_modules_6_modules_branch2_modules_1_modules_bn_buffers_running_mean_
+        l_self_modules_features_modules_6_modules_branch2_modules_1_modules_bn_buffers_running_var_ = L_self_modules_features_modules_6_modules_branch2_modules_1_modules_bn_buffers_running_var_
+        l_self_modules_features_modules_6_modules_branch2_modules_1_modules_bn_parameters_weight_ = L_self_modules_features_modules_6_modules_branch2_modules_1_modules_bn_parameters_weight_
+        l_self_modules_features_modules_6_modules_branch2_modules_1_modules_bn_parameters_bias_ = L_self_modules_features_modules_6_modules_branch2_modules_1_modules_bn_parameters_bias_
+        l_self_modules_features_modules_6_modules_branch2_modules_2_modules_conv_parameters_weight_ = L_self_modules_features_modules_6_modules_branch2_modules_2_modules_conv_parameters_weight_
+        l_self_modules_features_modules_6_modules_branch2_modules_2_modules_bn_buffers_running_mean_ = L_self_modules_features_modules_6_modules_branch2_modules_2_modules_bn_buffers_running_mean_
+        l_self_modules_features_modules_6_modules_branch2_modules_2_modules_bn_buffers_running_var_ = L_self_modules_features_modules_6_modules_branch2_modules_2_modules_bn_buffers_running_var_
+        l_self_modules_features_modules_6_modules_branch2_modules_2_modules_bn_parameters_weight_ = L_self_modules_features_modules_6_modules_branch2_modules_2_modules_bn_parameters_weight_
+        l_self_modules_features_modules_6_modules_branch2_modules_2_modules_bn_parameters_bias_ = L_self_modules_features_modules_6_modules_branch2_modules_2_modules_bn_parameters_bias_
+        l_self_modules_features_modules_6_modules_branch3_modules_1_modules_conv_parameters_weight_ = L_self_modules_features_modules_6_modules_branch3_modules_1_modules_conv_parameters_weight_
+        l_self_modules_features_modules_6_modules_branch3_modules_1_modules_bn_buffers_running_mean_ = L_self_modules_features_modules_6_modules_branch3_modules_1_modules_bn_buffers_running_mean_
+        l_self_modules_features_modules_6_modules_branch3_modules_1_modules_bn_buffers_running_var_ = L_self_modules_features_modules_6_modules_branch3_modules_1_modules_bn_buffers_running_var_
+        l_self_modules_features_modules_6_modules_branch3_modules_1_modules_bn_parameters_weight_ = L_self_modules_features_modules_6_modules_branch3_modules_1_modules_bn_parameters_weight_
+        l_self_modules_features_modules_6_modules_branch3_modules_1_modules_bn_parameters_bias_ = L_self_modules_features_modules_6_modules_branch3_modules_1_modules_bn_parameters_bias_
+        l_self_modules_features_modules_7_modules_branch0_modules_conv_parameters_weight_ = L_self_modules_features_modules_7_modules_branch0_modules_conv_parameters_weight_
+        l_self_modules_features_modules_7_modules_branch0_modules_bn_buffers_running_mean_ = L_self_modules_features_modules_7_modules_branch0_modules_bn_buffers_running_mean_
+        l_self_modules_features_modules_7_modules_branch0_modules_bn_buffers_running_var_ = L_self_modules_features_modules_7_modules_branch0_modules_bn_buffers_running_var_
+        l_self_modules_features_modules_7_modules_branch0_modules_bn_parameters_weight_ = L_self_modules_features_modules_7_modules_branch0_modules_bn_parameters_weight_
+        l_self_modules_features_modules_7_modules_branch0_modules_bn_parameters_bias_ = L_self_modules_features_modules_7_modules_branch0_modules_bn_parameters_bias_
+        l_self_modules_features_modules_7_modules_branch1_modules_0_modules_conv_parameters_weight_ = L_self_modules_features_modules_7_modules_branch1_modules_0_modules_conv_parameters_weight_
+        l_self_modules_features_modules_7_modules_branch1_modules_0_modules_bn_buffers_running_mean_ = L_self_modules_features_modules_7_modules_branch1_modules_0_modules_bn_buffers_running_mean_
+        l_self_modules_features_modules_7_modules_branch1_modules_0_modules_bn_buffers_running_var_ = L_self_modules_features_modules_7_modules_branch1_modules_0_modules_bn_buffers_running_var_
+        l_self_modules_features_modules_7_modules_branch1_modules_0_modules_bn_parameters_weight_ = L_self_modules_features_modules_7_modules_branch1_modules_0_modules_bn_parameters_weight_
+        l_self_modules_features_modules_7_modules_branch1_modules_0_modules_bn_parameters_bias_ = L_self_modules_features_modules_7_modules_branch1_modules_0_modules_bn_parameters_bias_
+        l_self_modules_features_modules_7_modules_branch1_modules_1_modules_conv_parameters_weight_ = L_self_modules_features_modules_7_modules_branch1_modules_1_modules_conv_parameters_weight_
+        l_self_modules_features_modules_7_modules_branch1_modules_1_modules_bn_buffers_running_mean_ = L_self_modules_features_modules_7_modules_branch1_modules_1_modules_bn_buffers_running_mean_
+        l_self_modules_features_modules_7_modules_branch1_modules_1_modules_bn_buffers_running_var_ = L_self_modules_features_modules_7_modules_branch1_modules_1_modules_bn_buffers_running_var_
+        l_self_modules_features_modules_7_modules_branch1_modules_1_modules_bn_parameters_weight_ = L_self_modules_features_modules_7_modules_branch1_modules_1_modules_bn_parameters_weight_
+        l_self_modules_features_modules_7_modules_branch1_modules_1_modules_bn_parameters_bias_ = L_self_modules_features_modules_7_modules_branch1_modules_1_modules_bn_parameters_bias_
+        l_self_modules_features_modules_7_modules_branch2_modules_0_modules_conv_parameters_weight_ = L_self_modules_features_modules_7_modules_branch2_modules_0_modules_conv_parameters_weight_
+        l_self_modules_features_modules_7_modules_branch2_modules_0_modules_bn_buffers_running_mean_ = L_self_modules_features_modules_7_modules_branch2_modules_0_modules_bn_buffers_running_mean_
+        l_self_modules_features_modules_7_modules_branch2_modules_0_modules_bn_buffers_running_var_ = L_self_modules_features_modules_7_modules_branch2_modules_0_modules_bn_buffers_running_var_
+        l_self_modules_features_modules_7_modules_branch2_modules_0_modules_bn_parameters_weight_ = L_self_modules_features_modules_7_modules_branch2_modules_0_modules_bn_parameters_weight_
+        l_self_modules_features_modules_7_modules_branch2_modules_0_modules_bn_parameters_bias_ = L_self_modules_features_modules_7_modules_branch2_modules_0_modules_bn_parameters_bias_
+        l_self_modules_features_modules_7_modules_branch2_modules_1_modules_conv_parameters_weight_ = L_self_modules_features_modules_7_modules_branch2_modules_1_modules_conv_parameters_weight_
+        l_self_modules_features_modules_7_modules_branch2_modules_1_modules_bn_buffers_running_mean_ = L_self_modules_features_modules_7_modules_branch2_modules_1_modules_bn_buffers_running_mean_
+        l_self_modules_features_modules_7_modules_branch2_modules_1_modules_bn_buffers_running_var_ = L_self_modules_features_modules_7_modules_branch2_modules_1_modules_bn_buffers_running_var_
+        l_self_modules_features_modules_7_modules_branch2_modules_1_modules_bn_parameters_weight_ = L_self_modules_features_modules_7_modules_branch2_modules_1_modules_bn_parameters_weight_
+        l_self_modules_features_modules_7_modules_branch2_modules_1_modules_bn_parameters_bias_ = L_self_modules_features_modules_7_modules_branch2_modules_1_modules_bn_parameters_bias_
+        l_self_modules_features_modules_7_modules_branch2_modules_2_modules_conv_parameters_weight_ = L_self_modules_features_modules_7_modules_branch2_modules_2_modules_conv_parameters_weight_
+        l_self_modules_features_modules_7_modules_branch2_modules_2_modules_bn_buffers_running_mean_ = L_self_modules_features_modules_7_modules_branch2_modules_2_modules_bn_buffers_running_mean_
+        l_self_modules_features_modules_7_modules_branch2_modules_2_modules_bn_buffers_running_var_ = L_self_modules_features_modules_7_modules_branch2_modules_2_modules_bn_buffers_running_var_
+        l_self_modules_features_modules_7_modules_branch2_modules_2_modules_bn_parameters_weight_ = L_self_modules_features_modules_7_modules_branch2_modules_2_modules_bn_parameters_weight_
+        l_self_modules_features_modules_7_modules_branch2_modules_2_modules_bn_parameters_bias_ = L_self_modules_features_modules_7_modules_branch2_modules_2_modules_bn_parameters_bias_
+        l_self_modules_features_modules_7_modules_branch3_modules_1_modules_conv_parameters_weight_ = L_self_modules_features_modules_7_modules_branch3_modules_1_modules_conv_parameters_weight_
+        l_self_modules_features_modules_7_modules_branch3_modules_1_modules_bn_buffers_running_mean_ = L_self_modules_features_modules_7_modules_branch3_modules_1_modules_bn_buffers_running_mean_
+        l_self_modules_features_modules_7_modules_branch3_modules_1_modules_bn_buffers_running_var_ = L_self_modules_features_modules_7_modules_branch3_modules_1_modules_bn_buffers_running_var_
+        l_self_modules_features_modules_7_modules_branch3_modules_1_modules_bn_parameters_weight_ = L_self_modules_features_modules_7_modules_branch3_modules_1_modules_bn_parameters_weight_
+        l_self_modules_features_modules_7_modules_branch3_modules_1_modules_bn_parameters_bias_ = L_self_modules_features_modules_7_modules_branch3_modules_1_modules_bn_parameters_bias_
+        l_self_modules_features_modules_8_modules_branch0_modules_conv_parameters_weight_ = L_self_modules_features_modules_8_modules_branch0_modules_conv_parameters_weight_
+        l_self_modules_features_modules_8_modules_branch0_modules_bn_buffers_running_mean_ = L_self_modules_features_modules_8_modules_branch0_modules_bn_buffers_running_mean_
+        l_self_modules_features_modules_8_modules_branch0_modules_bn_buffers_running_var_ = L_self_modules_features_modules_8_modules_branch0_modules_bn_buffers_running_var_
+        l_self_modules_features_modules_8_modules_branch0_modules_bn_parameters_weight_ = L_self_modules_features_modules_8_modules_branch0_modules_bn_parameters_weight_
+        l_self_modules_features_modules_8_modules_branch0_modules_bn_parameters_bias_ = L_self_modules_features_modules_8_modules_branch0_modules_bn_parameters_bias_
+        l_self_modules_features_modules_8_modules_branch1_modules_0_modules_conv_parameters_weight_ = L_self_modules_features_modules_8_modules_branch1_modules_0_modules_conv_parameters_weight_
+        l_self_modules_features_modules_8_modules_branch1_modules_0_modules_bn_buffers_running_mean_ = L_self_modules_features_modules_8_modules_branch1_modules_0_modules_bn_buffers_running_mean_
+        l_self_modules_features_modules_8_modules_branch1_modules_0_modules_bn_buffers_running_var_ = L_self_modules_features_modules_8_modules_branch1_modules_0_modules_bn_buffers_running_var_
+        l_self_modules_features_modules_8_modules_branch1_modules_0_modules_bn_parameters_weight_ = L_self_modules_features_modules_8_modules_branch1_modules_0_modules_bn_parameters_weight_
+        l_self_modules_features_modules_8_modules_branch1_modules_0_modules_bn_parameters_bias_ = L_self_modules_features_modules_8_modules_branch1_modules_0_modules_bn_parameters_bias_
+        l_self_modules_features_modules_8_modules_branch1_modules_1_modules_conv_parameters_weight_ = L_self_modules_features_modules_8_modules_branch1_modules_1_modules_conv_parameters_weight_
+        l_self_modules_features_modules_8_modules_branch1_modules_1_modules_bn_buffers_running_mean_ = L_self_modules_features_modules_8_modules_branch1_modules_1_modules_bn_buffers_running_mean_
+        l_self_modules_features_modules_8_modules_branch1_modules_1_modules_bn_buffers_running_var_ = L_self_modules_features_modules_8_modules_branch1_modules_1_modules_bn_buffers_running_var_
+        l_self_modules_features_modules_8_modules_branch1_modules_1_modules_bn_parameters_weight_ = L_self_modules_features_modules_8_modules_branch1_modules_1_modules_bn_parameters_weight_
+        l_self_modules_features_modules_8_modules_branch1_modules_1_modules_bn_parameters_bias_ = L_self_modules_features_modules_8_modules_branch1_modules_1_modules_bn_parameters_bias_
+        l_self_modules_features_modules_8_modules_branch2_modules_0_modules_conv_parameters_weight_ = L_self_modules_features_modules_8_modules_branch2_modules_0_modules_conv_parameters_weight_
+        l_self_modules_features_modules_8_modules_branch2_modules_0_modules_bn_buffers_running_mean_ = L_self_modules_features_modules_8_modules_branch2_modules_0_modules_bn_buffers_running_mean_
+        l_self_modules_features_modules_8_modules_branch2_modules_0_modules_bn_buffers_running_var_ = L_self_modules_features_modules_8_modules_branch2_modules_0_modules_bn_buffers_running_var_
+        l_self_modules_features_modules_8_modules_branch2_modules_0_modules_bn_parameters_weight_ = L_self_modules_features_modules_8_modules_branch2_modules_0_modules_bn_parameters_weight_
+        l_self_modules_features_modules_8_modules_branch2_modules_0_modules_bn_parameters_bias_ = L_self_modules_features_modules_8_modules_branch2_modules_0_modules_bn_parameters_bias_
+        l_self_modules_features_modules_8_modules_branch2_modules_1_modules_conv_parameters_weight_ = L_self_modules_features_modules_8_modules_branch2_modules_1_modules_conv_parameters_weight_
+        l_self_modules_features_modules_8_modules_branch2_modules_1_modules_bn_buffers_running_mean_ = L_self_modules_features_modules_8_modules_branch2_modules_1_modules_bn_buffers_running_mean_
+        l_self_modules_features_modules_8_modules_branch2_modules_1_modules_bn_buffers_running_var_ = L_self_modules_features_modules_8_modules_branch2_modules_1_modules_bn_buffers_running_var_
+        l_self_modules_features_modules_8_modules_branch2_modules_1_modules_bn_parameters_weight_ = L_self_modules_features_modules_8_modules_branch2_modules_1_modules_bn_parameters_weight_
+        l_self_modules_features_modules_8_modules_branch2_modules_1_modules_bn_parameters_bias_ = L_self_modules_features_modules_8_modules_branch2_modules_1_modules_bn_parameters_bias_
+        l_self_modules_features_modules_8_modules_branch2_modules_2_modules_conv_parameters_weight_ = L_self_modules_features_modules_8_modules_branch2_modules_2_modules_conv_parameters_weight_
+        l_self_modules_features_modules_8_modules_branch2_modules_2_modules_bn_buffers_running_mean_ = L_self_modules_features_modules_8_modules_branch2_modules_2_modules_bn_buffers_running_mean_
+        l_self_modules_features_modules_8_modules_branch2_modules_2_modules_bn_buffers_running_var_ = L_self_modules_features_modules_8_modules_branch2_modules_2_modules_bn_buffers_running_var_
+        l_self_modules_features_modules_8_modules_branch2_modules_2_modules_bn_parameters_weight_ = L_self_modules_features_modules_8_modules_branch2_modules_2_modules_bn_parameters_weight_
+        l_self_modules_features_modules_8_modules_branch2_modules_2_modules_bn_parameters_bias_ = L_self_modules_features_modules_8_modules_branch2_modules_2_modules_bn_parameters_bias_
+        l_self_modules_features_modules_8_modules_branch3_modules_1_modules_conv_parameters_weight_ = L_self_modules_features_modules_8_modules_branch3_modules_1_modules_conv_parameters_weight_
+        l_self_modules_features_modules_8_modules_branch3_modules_1_modules_bn_buffers_running_mean_ = L_self_modules_features_modules_8_modules_branch3_modules_1_modules_bn_buffers_running_mean_
+        l_self_modules_features_modules_8_modules_branch3_modules_1_modules_bn_buffers_running_var_ = L_self_modules_features_modules_8_modules_branch3_modules_1_modules_bn_buffers_running_var_
+        l_self_modules_features_modules_8_modules_branch3_modules_1_modules_bn_parameters_weight_ = L_self_modules_features_modules_8_modules_branch3_modules_1_modules_bn_parameters_weight_
+        l_self_modules_features_modules_8_modules_branch3_modules_1_modules_bn_parameters_bias_ = L_self_modules_features_modules_8_modules_branch3_modules_1_modules_bn_parameters_bias_
+        l_self_modules_features_modules_9_modules_branch0_modules_conv_parameters_weight_ = L_self_modules_features_modules_9_modules_branch0_modules_conv_parameters_weight_
+        l_self_modules_features_modules_9_modules_branch0_modules_bn_buffers_running_mean_ = L_self_modules_features_modules_9_modules_branch0_modules_bn_buffers_running_mean_
+        l_self_modules_features_modules_9_modules_branch0_modules_bn_buffers_running_var_ = L_self_modules_features_modules_9_modules_branch0_modules_bn_buffers_running_var_
+        l_self_modules_features_modules_9_modules_branch0_modules_bn_parameters_weight_ = L_self_modules_features_modules_9_modules_branch0_modules_bn_parameters_weight_
+        l_self_modules_features_modules_9_modules_branch0_modules_bn_parameters_bias_ = L_self_modules_features_modules_9_modules_branch0_modules_bn_parameters_bias_
+        l_self_modules_features_modules_9_modules_branch1_modules_0_modules_conv_parameters_weight_ = L_self_modules_features_modules_9_modules_branch1_modules_0_modules_conv_parameters_weight_
+        l_self_modules_features_modules_9_modules_branch1_modules_0_modules_bn_buffers_running_mean_ = L_self_modules_features_modules_9_modules_branch1_modules_0_modules_bn_buffers_running_mean_
+        l_self_modules_features_modules_9_modules_branch1_modules_0_modules_bn_buffers_running_var_ = L_self_modules_features_modules_9_modules_branch1_modules_0_modules_bn_buffers_running_var_
+        l_self_modules_features_modules_9_modules_branch1_modules_0_modules_bn_parameters_weight_ = L_self_modules_features_modules_9_modules_branch1_modules_0_modules_bn_parameters_weight_
+        l_self_modules_features_modules_9_modules_branch1_modules_0_modules_bn_parameters_bias_ = L_self_modules_features_modules_9_modules_branch1_modules_0_modules_bn_parameters_bias_
+        l_self_modules_features_modules_9_modules_branch1_modules_1_modules_conv_parameters_weight_ = L_self_modules_features_modules_9_modules_branch1_modules_1_modules_conv_parameters_weight_
+        l_self_modules_features_modules_9_modules_branch1_modules_1_modules_bn_buffers_running_mean_ = L_self_modules_features_modules_9_modules_branch1_modules_1_modules_bn_buffers_running_mean_
+        l_self_modules_features_modules_9_modules_branch1_modules_1_modules_bn_buffers_running_var_ = L_self_modules_features_modules_9_modules_branch1_modules_1_modules_bn_buffers_running_var_
+        l_self_modules_features_modules_9_modules_branch1_modules_1_modules_bn_parameters_weight_ = L_self_modules_features_modules_9_modules_branch1_modules_1_modules_bn_parameters_weight_
+        l_self_modules_features_modules_9_modules_branch1_modules_1_modules_bn_parameters_bias_ = L_self_modules_features_modules_9_modules_branch1_modules_1_modules_bn_parameters_bias_
+        l_self_modules_features_modules_9_modules_branch2_modules_0_modules_conv_parameters_weight_ = L_self_modules_features_modules_9_modules_branch2_modules_0_modules_conv_parameters_weight_
+        l_self_modules_features_modules_9_modules_branch2_modules_0_modules_bn_buffers_running_mean_ = L_self_modules_features_modules_9_modules_branch2_modules_0_modules_bn_buffers_running_mean_
+        l_self_modules_features_modules_9_modules_branch2_modules_0_modules_bn_buffers_running_var_ = L_self_modules_features_modules_9_modules_branch2_modules_0_modules_bn_buffers_running_var_
+        l_self_modules_features_modules_9_modules_branch2_modules_0_modules_bn_parameters_weight_ = L_self_modules_features_modules_9_modules_branch2_modules_0_modules_bn_parameters_weight_
+        l_self_modules_features_modules_9_modules_branch2_modules_0_modules_bn_parameters_bias_ = L_self_modules_features_modules_9_modules_branch2_modules_0_modules_bn_parameters_bias_
+        l_self_modules_features_modules_9_modules_branch2_modules_1_modules_conv_parameters_weight_ = L_self_modules_features_modules_9_modules_branch2_modules_1_modules_conv_parameters_weight_
+        l_self_modules_features_modules_9_modules_branch2_modules_1_modules_bn_buffers_running_mean_ = L_self_modules_features_modules_9_modules_branch2_modules_1_modules_bn_buffers_running_mean_
+        l_self_modules_features_modules_9_modules_branch2_modules_1_modules_bn_buffers_running_var_ = L_self_modules_features_modules_9_modules_branch2_modules_1_modules_bn_buffers_running_var_
+        l_self_modules_features_modules_9_modules_branch2_modules_1_modules_bn_parameters_weight_ = L_self_modules_features_modules_9_modules_branch2_modules_1_modules_bn_parameters_weight_
+        l_self_modules_features_modules_9_modules_branch2_modules_1_modules_bn_parameters_bias_ = L_self_modules_features_modules_9_modules_branch2_modules_1_modules_bn_parameters_bias_
+        l_self_modules_features_modules_9_modules_branch2_modules_2_modules_conv_parameters_weight_ = L_self_modules_features_modules_9_modules_branch2_modules_2_modules_conv_parameters_weight_
+        l_self_modules_features_modules_9_modules_branch2_modules_2_modules_bn_buffers_running_mean_ = L_self_modules_features_modules_9_modules_branch2_modules_2_modules_bn_buffers_running_mean_
+        l_self_modules_features_modules_9_modules_branch2_modules_2_modules_bn_buffers_running_var_ = L_self_modules_features_modules_9_modules_branch2_modules_2_modules_bn_buffers_running_var_
+        l_self_modules_features_modules_9_modules_branch2_modules_2_modules_bn_parameters_weight_ = L_self_modules_features_modules_9_modules_branch2_modules_2_modules_bn_parameters_weight_
+        l_self_modules_features_modules_9_modules_branch2_modules_2_modules_bn_parameters_bias_ = L_self_modules_features_modules_9_modules_branch2_modules_2_modules_bn_parameters_bias_
+        l_self_modules_features_modules_9_modules_branch3_modules_1_modules_conv_parameters_weight_ = L_self_modules_features_modules_9_modules_branch3_modules_1_modules_conv_parameters_weight_
+        l_self_modules_features_modules_9_modules_branch3_modules_1_modules_bn_buffers_running_mean_ = L_self_modules_features_modules_9_modules_branch3_modules_1_modules_bn_buffers_running_mean_
+        l_self_modules_features_modules_9_modules_branch3_modules_1_modules_bn_buffers_running_var_ = L_self_modules_features_modules_9_modules_branch3_modules_1_modules_bn_buffers_running_var_
+        l_self_modules_features_modules_9_modules_branch3_modules_1_modules_bn_parameters_weight_ = L_self_modules_features_modules_9_modules_branch3_modules_1_modules_bn_parameters_weight_
+        l_self_modules_features_modules_9_modules_branch3_modules_1_modules_bn_parameters_bias_ = L_self_modules_features_modules_9_modules_branch3_modules_1_modules_bn_parameters_bias_
+        l_self_modules_features_modules_10_modules_branch0_modules_conv_parameters_weight_ = L_self_modules_features_modules_10_modules_branch0_modules_conv_parameters_weight_
+        l_self_modules_features_modules_10_modules_branch0_modules_bn_buffers_running_mean_ = L_self_modules_features_modules_10_modules_branch0_modules_bn_buffers_running_mean_
+        l_self_modules_features_modules_10_modules_branch0_modules_bn_buffers_running_var_ = L_self_modules_features_modules_10_modules_branch0_modules_bn_buffers_running_var_
+        l_self_modules_features_modules_10_modules_branch0_modules_bn_parameters_weight_ = L_self_modules_features_modules_10_modules_branch0_modules_bn_parameters_weight_
+        l_self_modules_features_modules_10_modules_branch0_modules_bn_parameters_bias_ = L_self_modules_features_modules_10_modules_branch0_modules_bn_parameters_bias_
+        l_self_modules_features_modules_10_modules_branch1_modules_0_modules_conv_parameters_weight_ = L_self_modules_features_modules_10_modules_branch1_modules_0_modules_conv_parameters_weight_
+        l_self_modules_features_modules_10_modules_branch1_modules_0_modules_bn_buffers_running_mean_ = L_self_modules_features_modules_10_modules_branch1_modules_0_modules_bn_buffers_running_mean_
+        l_self_modules_features_modules_10_modules_branch1_modules_0_modules_bn_buffers_running_var_ = L_self_modules_features_modules_10_modules_branch1_modules_0_modules_bn_buffers_running_var_
+        l_self_modules_features_modules_10_modules_branch1_modules_0_modules_bn_parameters_weight_ = L_self_modules_features_modules_10_modules_branch1_modules_0_modules_bn_parameters_weight_
+        l_self_modules_features_modules_10_modules_branch1_modules_0_modules_bn_parameters_bias_ = L_self_modules_features_modules_10_modules_branch1_modules_0_modules_bn_parameters_bias_
+        l_self_modules_features_modules_10_modules_branch1_modules_1_modules_conv_parameters_weight_ = L_self_modules_features_modules_10_modules_branch1_modules_1_modules_conv_parameters_weight_
+        l_self_modules_features_modules_10_modules_branch1_modules_1_modules_bn_buffers_running_mean_ = L_self_modules_features_modules_10_modules_branch1_modules_1_modules_bn_buffers_running_mean_
+        l_self_modules_features_modules_10_modules_branch1_modules_1_modules_bn_buffers_running_var_ = L_self_modules_features_modules_10_modules_branch1_modules_1_modules_bn_buffers_running_var_
+        l_self_modules_features_modules_10_modules_branch1_modules_1_modules_bn_parameters_weight_ = L_self_modules_features_modules_10_modules_branch1_modules_1_modules_bn_parameters_weight_
+        l_self_modules_features_modules_10_modules_branch1_modules_1_modules_bn_parameters_bias_ = L_self_modules_features_modules_10_modules_branch1_modules_1_modules_bn_parameters_bias_
+        l_self_modules_features_modules_10_modules_branch1_modules_2_modules_conv_parameters_weight_ = L_self_modules_features_modules_10_modules_branch1_modules_2_modules_conv_parameters_weight_
+        l_self_modules_features_modules_10_modules_branch1_modules_2_modules_bn_buffers_running_mean_ = L_self_modules_features_modules_10_modules_branch1_modules_2_modules_bn_buffers_running_mean_
+        l_self_modules_features_modules_10_modules_branch1_modules_2_modules_bn_buffers_running_var_ = L_self_modules_features_modules_10_modules_branch1_modules_2_modules_bn_buffers_running_var_
+        l_self_modules_features_modules_10_modules_branch1_modules_2_modules_bn_parameters_weight_ = L_self_modules_features_modules_10_modules_branch1_modules_2_modules_bn_parameters_weight_
+        l_self_modules_features_modules_10_modules_branch1_modules_2_modules_bn_parameters_bias_ = L_self_modules_features_modules_10_modules_branch1_modules_2_modules_bn_parameters_bias_
+        l_self_modules_features_modules_11_modules_branch0_modules_conv_parameters_weight_ = L_self_modules_features_modules_11_modules_branch0_modules_conv_parameters_weight_
+        l_self_modules_features_modules_11_modules_branch0_modules_bn_buffers_running_mean_ = L_self_modules_features_modules_11_modules_branch0_modules_bn_buffers_running_mean_
+        l_self_modules_features_modules_11_modules_branch0_modules_bn_buffers_running_var_ = L_self_modules_features_modules_11_modules_branch0_modules_bn_buffers_running_var_
+        l_self_modules_features_modules_11_modules_branch0_modules_bn_parameters_weight_ = L_self_modules_features_modules_11_modules_branch0_modules_bn_parameters_weight_
+        l_self_modules_features_modules_11_modules_branch0_modules_bn_parameters_bias_ = L_self_modules_features_modules_11_modules_branch0_modules_bn_parameters_bias_
+        l_self_modules_features_modules_11_modules_branch1_modules_0_modules_conv_parameters_weight_ = L_self_modules_features_modules_11_modules_branch1_modules_0_modules_conv_parameters_weight_
+        l_self_modules_features_modules_11_modules_branch1_modules_0_modules_bn_buffers_running_mean_ = L_self_modules_features_modules_11_modules_branch1_modules_0_modules_bn_buffers_running_mean_
+        l_self_modules_features_modules_11_modules_branch1_modules_0_modules_bn_buffers_running_var_ = L_self_modules_features_modules_11_modules_branch1_modules_0_modules_bn_buffers_running_var_
+        l_self_modules_features_modules_11_modules_branch1_modules_0_modules_bn_parameters_weight_ = L_self_modules_features_modules_11_modules_branch1_modules_0_modules_bn_parameters_weight_
+        l_self_modules_features_modules_11_modules_branch1_modules_0_modules_bn_parameters_bias_ = L_self_modules_features_modules_11_modules_branch1_modules_0_modules_bn_parameters_bias_
+        l_self_modules_features_modules_11_modules_branch1_modules_1_modules_conv_parameters_weight_ = L_self_modules_features_modules_11_modules_branch1_modules_1_modules_conv_parameters_weight_
+        l_self_modules_features_modules_11_modules_branch1_modules_1_modules_bn_buffers_running_mean_ = L_self_modules_features_modules_11_modules_branch1_modules_1_modules_bn_buffers_running_mean_
+        l_self_modules_features_modules_11_modules_branch1_modules_1_modules_bn_buffers_running_var_ = L_self_modules_features_modules_11_modules_branch1_modules_1_modules_bn_buffers_running_var_
+        l_self_modules_features_modules_11_modules_branch1_modules_1_modules_bn_parameters_weight_ = L_self_modules_features_modules_11_modules_branch1_modules_1_modules_bn_parameters_weight_
+        l_self_modules_features_modules_11_modules_branch1_modules_1_modules_bn_parameters_bias_ = L_self_modules_features_modules_11_modules_branch1_modules_1_modules_bn_parameters_bias_
+        l_self_modules_features_modules_11_modules_branch1_modules_2_modules_conv_parameters_weight_ = L_self_modules_features_modules_11_modules_branch1_modules_2_modules_conv_parameters_weight_
+        l_self_modules_features_modules_11_modules_branch1_modules_2_modules_bn_buffers_running_mean_ = L_self_modules_features_modules_11_modules_branch1_modules_2_modules_bn_buffers_running_mean_
+        l_self_modules_features_modules_11_modules_branch1_modules_2_modules_bn_buffers_running_var_ = L_self_modules_features_modules_11_modules_branch1_modules_2_modules_bn_buffers_running_var_
+        l_self_modules_features_modules_11_modules_branch1_modules_2_modules_bn_parameters_weight_ = L_self_modules_features_modules_11_modules_branch1_modules_2_modules_bn_parameters_weight_
+        l_self_modules_features_modules_11_modules_branch1_modules_2_modules_bn_parameters_bias_ = L_self_modules_features_modules_11_modules_branch1_modules_2_modules_bn_parameters_bias_
+        l_self_modules_features_modules_11_modules_branch2_modules_0_modules_conv_parameters_weight_ = L_self_modules_features_modules_11_modules_branch2_modules_0_modules_conv_parameters_weight_
+        l_self_modules_features_modules_11_modules_branch2_modules_0_modules_bn_buffers_running_mean_ = L_self_modules_features_modules_11_modules_branch2_modules_0_modules_bn_buffers_running_mean_
+        l_self_modules_features_modules_11_modules_branch2_modules_0_modules_bn_buffers_running_var_ = L_self_modules_features_modules_11_modules_branch2_modules_0_modules_bn_buffers_running_var_
+        l_self_modules_features_modules_11_modules_branch2_modules_0_modules_bn_parameters_weight_ = L_self_modules_features_modules_11_modules_branch2_modules_0_modules_bn_parameters_weight_
+        l_self_modules_features_modules_11_modules_branch2_modules_0_modules_bn_parameters_bias_ = L_self_modules_features_modules_11_modules_branch2_modules_0_modules_bn_parameters_bias_
+        l_self_modules_features_modules_11_modules_branch2_modules_1_modules_conv_parameters_weight_ = L_self_modules_features_modules_11_modules_branch2_modules_1_modules_conv_parameters_weight_
+        l_self_modules_features_modules_11_modules_branch2_modules_1_modules_bn_buffers_running_mean_ = L_self_modules_features_modules_11_modules_branch2_modules_1_modules_bn_buffers_running_mean_
+        l_self_modules_features_modules_11_modules_branch2_modules_1_modules_bn_buffers_running_var_ = L_self_modules_features_modules_11_modules_branch2_modules_1_modules_bn_buffers_running_var_
+        l_self_modules_features_modules_11_modules_branch2_modules_1_modules_bn_parameters_weight_ = L_self_modules_features_modules_11_modules_branch2_modules_1_modules_bn_parameters_weight_
+        l_self_modules_features_modules_11_modules_branch2_modules_1_modules_bn_parameters_bias_ = L_self_modules_features_modules_11_modules_branch2_modules_1_modules_bn_parameters_bias_
+        l_self_modules_features_modules_11_modules_branch2_modules_2_modules_conv_parameters_weight_ = L_self_modules_features_modules_11_modules_branch2_modules_2_modules_conv_parameters_weight_
+        l_self_modules_features_modules_11_modules_branch2_modules_2_modules_bn_buffers_running_mean_ = L_self_modules_features_modules_11_modules_branch2_modules_2_modules_bn_buffers_running_mean_
+        l_self_modules_features_modules_11_modules_branch2_modules_2_modules_bn_buffers_running_var_ = L_self_modules_features_modules_11_modules_branch2_modules_2_modules_bn_buffers_running_var_
+        l_self_modules_features_modules_11_modules_branch2_modules_2_modules_bn_parameters_weight_ = L_self_modules_features_modules_11_modules_branch2_modules_2_modules_bn_parameters_weight_
+        l_self_modules_features_modules_11_modules_branch2_modules_2_modules_bn_parameters_bias_ = L_self_modules_features_modules_11_modules_branch2_modules_2_modules_bn_parameters_bias_
+        l_self_modules_features_modules_11_modules_branch2_modules_3_modules_conv_parameters_weight_ = L_self_modules_features_modules_11_modules_branch2_modules_3_modules_conv_parameters_weight_
+        l_self_modules_features_modules_11_modules_branch2_modules_3_modules_bn_buffers_running_mean_ = L_self_modules_features_modules_11_modules_branch2_modules_3_modules_bn_buffers_running_mean_
+        l_self_modules_features_modules_11_modules_branch2_modules_3_modules_bn_buffers_running_var_ = L_self_modules_features_modules_11_modules_branch2_modules_3_modules_bn_buffers_running_var_
+        l_self_modules_features_modules_11_modules_branch2_modules_3_modules_bn_parameters_weight_ = L_self_modules_features_modules_11_modules_branch2_modules_3_modules_bn_parameters_weight_
+        l_self_modules_features_modules_11_modules_branch2_modules_3_modules_bn_parameters_bias_ = L_self_modules_features_modules_11_modules_branch2_modules_3_modules_bn_parameters_bias_
+        l_self_modules_features_modules_11_modules_branch2_modules_4_modules_conv_parameters_weight_ = L_self_modules_features_modules_11_modules_branch2_modules_4_modules_conv_parameters_weight_
+        l_self_modules_features_modules_11_modules_branch2_modules_4_modules_bn_buffers_running_mean_ = L_self_modules_features_modules_11_modules_branch2_modules_4_modules_bn_buffers_running_mean_
+        l_self_modules_features_modules_11_modules_branch2_modules_4_modules_bn_buffers_running_var_ = L_self_modules_features_modules_11_modules_branch2_modules_4_modules_bn_buffers_running_var_
+        l_self_modules_features_modules_11_modules_branch2_modules_4_modules_bn_parameters_weight_ = L_self_modules_features_modules_11_modules_branch2_modules_4_modules_bn_parameters_weight_
+        l_self_modules_features_modules_11_modules_branch2_modules_4_modules_bn_parameters_bias_ = L_self_modules_features_modules_11_modules_branch2_modules_4_modules_bn_parameters_bias_
+        l_self_modules_features_modules_11_modules_branch3_modules_1_modules_conv_parameters_weight_ = L_self_modules_features_modules_11_modules_branch3_modules_1_modules_conv_parameters_weight_
+        l_self_modules_features_modules_11_modules_branch3_modules_1_modules_bn_buffers_running_mean_ = L_self_modules_features_modules_11_modules_branch3_modules_1_modules_bn_buffers_running_mean_
+        l_self_modules_features_modules_11_modules_branch3_modules_1_modules_bn_buffers_running_var_ = L_self_modules_features_modules_11_modules_branch3_modules_1_modules_bn_buffers_running_var_
+        l_self_modules_features_modules_11_modules_branch3_modules_1_modules_bn_parameters_weight_ = L_self_modules_features_modules_11_modules_branch3_modules_1_modules_bn_parameters_weight_
+        l_self_modules_features_modules_11_modules_branch3_modules_1_modules_bn_parameters_bias_ = L_self_modules_features_modules_11_modules_branch3_modules_1_modules_bn_parameters_bias_
+        l_self_modules_features_modules_12_modules_branch0_modules_conv_parameters_weight_ = L_self_modules_features_modules_12_modules_branch0_modules_conv_parameters_weight_
+        l_self_modules_features_modules_12_modules_branch0_modules_bn_buffers_running_mean_ = L_self_modules_features_modules_12_modules_branch0_modules_bn_buffers_running_mean_
+        l_self_modules_features_modules_12_modules_branch0_modules_bn_buffers_running_var_ = L_self_modules_features_modules_12_modules_branch0_modules_bn_buffers_running_var_
+        l_self_modules_features_modules_12_modules_branch0_modules_bn_parameters_weight_ = L_self_modules_features_modules_12_modules_branch0_modules_bn_parameters_weight_
+        l_self_modules_features_modules_12_modules_branch0_modules_bn_parameters_bias_ = L_self_modules_features_modules_12_modules_branch0_modules_bn_parameters_bias_
+        l_self_modules_features_modules_12_modules_branch1_modules_0_modules_conv_parameters_weight_ = L_self_modules_features_modules_12_modules_branch1_modules_0_modules_conv_parameters_weight_
+        l_self_modules_features_modules_12_modules_branch1_modules_0_modules_bn_buffers_running_mean_ = L_self_modules_features_modules_12_modules_branch1_modules_0_modules_bn_buffers_running_mean_
+        l_self_modules_features_modules_12_modules_branch1_modules_0_modules_bn_buffers_running_var_ = L_self_modules_features_modules_12_modules_branch1_modules_0_modules_bn_buffers_running_var_
+        l_self_modules_features_modules_12_modules_branch1_modules_0_modules_bn_parameters_weight_ = L_self_modules_features_modules_12_modules_branch1_modules_0_modules_bn_parameters_weight_
+        l_self_modules_features_modules_12_modules_branch1_modules_0_modules_bn_parameters_bias_ = L_self_modules_features_modules_12_modules_branch1_modules_0_modules_bn_parameters_bias_
+        l_self_modules_features_modules_12_modules_branch1_modules_1_modules_conv_parameters_weight_ = L_self_modules_features_modules_12_modules_branch1_modules_1_modules_conv_parameters_weight_
+        l_self_modules_features_modules_12_modules_branch1_modules_1_modules_bn_buffers_running_mean_ = L_self_modules_features_modules_12_modules_branch1_modules_1_modules_bn_buffers_running_mean_
+        l_self_modules_features_modules_12_modules_branch1_modules_1_modules_bn_buffers_running_var_ = L_self_modules_features_modules_12_modules_branch1_modules_1_modules_bn_buffers_running_var_
+        l_self_modules_features_modules_12_modules_branch1_modules_1_modules_bn_parameters_weight_ = L_self_modules_features_modules_12_modules_branch1_modules_1_modules_bn_parameters_weight_
+        l_self_modules_features_modules_12_modules_branch1_modules_1_modules_bn_parameters_bias_ = L_self_modules_features_modules_12_modules_branch1_modules_1_modules_bn_parameters_bias_
+        l_self_modules_features_modules_12_modules_branch1_modules_2_modules_conv_parameters_weight_ = L_self_modules_features_modules_12_modules_branch1_modules_2_modules_conv_parameters_weight_
+        l_self_modules_features_modules_12_modules_branch1_modules_2_modules_bn_buffers_running_mean_ = L_self_modules_features_modules_12_modules_branch1_modules_2_modules_bn_buffers_running_mean_
+        l_self_modules_features_modules_12_modules_branch1_modules_2_modules_bn_buffers_running_var_ = L_self_modules_features_modules_12_modules_branch1_modules_2_modules_bn_buffers_running_var_
+        l_self_modules_features_modules_12_modules_branch1_modules_2_modules_bn_parameters_weight_ = L_self_modules_features_modules_12_modules_branch1_modules_2_modules_bn_parameters_weight_
+        l_self_modules_features_modules_12_modules_branch1_modules_2_modules_bn_parameters_bias_ = L_self_modules_features_modules_12_modules_branch1_modules_2_modules_bn_parameters_bias_
+        l_self_modules_features_modules_12_modules_branch2_modules_0_modules_conv_parameters_weight_ = L_self_modules_features_modules_12_modules_branch2_modules_0_modules_conv_parameters_weight_
+        l_self_modules_features_modules_12_modules_branch2_modules_0_modules_bn_buffers_running_mean_ = L_self_modules_features_modules_12_modules_branch2_modules_0_modules_bn_buffers_running_mean_
+        l_self_modules_features_modules_12_modules_branch2_modules_0_modules_bn_buffers_running_var_ = L_self_modules_features_modules_12_modules_branch2_modules_0_modules_bn_buffers_running_var_
+        l_self_modules_features_modules_12_modules_branch2_modules_0_modules_bn_parameters_weight_ = L_self_modules_features_modules_12_modules_branch2_modules_0_modules_bn_parameters_weight_
+        l_self_modules_features_modules_12_modules_branch2_modules_0_modules_bn_parameters_bias_ = L_self_modules_features_modules_12_modules_branch2_modules_0_modules_bn_parameters_bias_
+        l_self_modules_features_modules_12_modules_branch2_modules_1_modules_conv_parameters_weight_ = L_self_modules_features_modules_12_modules_branch2_modules_1_modules_conv_parameters_weight_
+        l_self_modules_features_modules_12_modules_branch2_modules_1_modules_bn_buffers_running_mean_ = L_self_modules_features_modules_12_modules_branch2_modules_1_modules_bn_buffers_running_mean_
+        l_self_modules_features_modules_12_modules_branch2_modules_1_modules_bn_buffers_running_var_ = L_self_modules_features_modules_12_modules_branch2_modules_1_modules_bn_buffers_running_var_
+        l_self_modules_features_modules_12_modules_branch2_modules_1_modules_bn_parameters_weight_ = L_self_modules_features_modules_12_modules_branch2_modules_1_modules_bn_parameters_weight_
+        l_self_modules_features_modules_12_modules_branch2_modules_1_modules_bn_parameters_bias_ = L_self_modules_features_modules_12_modules_branch2_modules_1_modules_bn_parameters_bias_
+        l_self_modules_features_modules_12_modules_branch2_modules_2_modules_conv_parameters_weight_ = L_self_modules_features_modules_12_modules_branch2_modules_2_modules_conv_parameters_weight_
+        l_self_modules_features_modules_12_modules_branch2_modules_2_modules_bn_buffers_running_mean_ = L_self_modules_features_modules_12_modules_branch2_modules_2_modules_bn_buffers_running_mean_
+        l_self_modules_features_modules_12_modules_branch2_modules_2_modules_bn_buffers_running_var_ = L_self_modules_features_modules_12_modules_branch2_modules_2_modules_bn_buffers_running_var_
+        l_self_modules_features_modules_12_modules_branch2_modules_2_modules_bn_parameters_weight_ = L_self_modules_features_modules_12_modules_branch2_modules_2_modules_bn_parameters_weight_
+        l_self_modules_features_modules_12_modules_branch2_modules_2_modules_bn_parameters_bias_ = L_self_modules_features_modules_12_modules_branch2_modules_2_modules_bn_parameters_bias_
+        l_self_modules_features_modules_12_modules_branch2_modules_3_modules_conv_parameters_weight_ = L_self_modules_features_modules_12_modules_branch2_modules_3_modules_conv_parameters_weight_
+        l_self_modules_features_modules_12_modules_branch2_modules_3_modules_bn_buffers_running_mean_ = L_self_modules_features_modules_12_modules_branch2_modules_3_modules_bn_buffers_running_mean_
+        l_self_modules_features_modules_12_modules_branch2_modules_3_modules_bn_buffers_running_var_ = L_self_modules_features_modules_12_modules_branch2_modules_3_modules_bn_buffers_running_var_
+        l_self_modules_features_modules_12_modules_branch2_modules_3_modules_bn_parameters_weight_ = L_self_modules_features_modules_12_modules_branch2_modules_3_modules_bn_parameters_weight_
+        l_self_modules_features_modules_12_modules_branch2_modules_3_modules_bn_parameters_bias_ = L_self_modules_features_modules_12_modules_branch2_modules_3_modules_bn_parameters_bias_
+        l_self_modules_features_modules_12_modules_branch2_modules_4_modules_conv_parameters_weight_ = L_self_modules_features_modules_12_modules_branch2_modules_4_modules_conv_parameters_weight_
+        l_self_modules_features_modules_12_modules_branch2_modules_4_modules_bn_buffers_running_mean_ = L_self_modules_features_modules_12_modules_branch2_modules_4_modules_bn_buffers_running_mean_
+        l_self_modules_features_modules_12_modules_branch2_modules_4_modules_bn_buffers_running_var_ = L_self_modules_features_modules_12_modules_branch2_modules_4_modules_bn_buffers_running_var_
+        l_self_modules_features_modules_12_modules_branch2_modules_4_modules_bn_parameters_weight_ = L_self_modules_features_modules_12_modules_branch2_modules_4_modules_bn_parameters_weight_
+        l_self_modules_features_modules_12_modules_branch2_modules_4_modules_bn_parameters_bias_ = L_self_modules_features_modules_12_modules_branch2_modules_4_modules_bn_parameters_bias_
+        l_self_modules_features_modules_12_modules_branch3_modules_1_modules_conv_parameters_weight_ = L_self_modules_features_modules_12_modules_branch3_modules_1_modules_conv_parameters_weight_
+        l_self_modules_features_modules_12_modules_branch3_modules_1_modules_bn_buffers_running_mean_ = L_self_modules_features_modules_12_modules_branch3_modules_1_modules_bn_buffers_running_mean_
+        l_self_modules_features_modules_12_modules_branch3_modules_1_modules_bn_buffers_running_var_ = L_self_modules_features_modules_12_modules_branch3_modules_1_modules_bn_buffers_running_var_
+        l_self_modules_features_modules_12_modules_branch3_modules_1_modules_bn_parameters_weight_ = L_self_modules_features_modules_12_modules_branch3_modules_1_modules_bn_parameters_weight_
+        l_self_modules_features_modules_12_modules_branch3_modules_1_modules_bn_parameters_bias_ = L_self_modules_features_modules_12_modules_branch3_modules_1_modules_bn_parameters_bias_
+        l_self_modules_features_modules_13_modules_branch0_modules_conv_parameters_weight_ = L_self_modules_features_modules_13_modules_branch0_modules_conv_parameters_weight_
+        l_self_modules_features_modules_13_modules_branch0_modules_bn_buffers_running_mean_ = L_self_modules_features_modules_13_modules_branch0_modules_bn_buffers_running_mean_
+        l_self_modules_features_modules_13_modules_branch0_modules_bn_buffers_running_var_ = L_self_modules_features_modules_13_modules_branch0_modules_bn_buffers_running_var_
+        l_self_modules_features_modules_13_modules_branch0_modules_bn_parameters_weight_ = L_self_modules_features_modules_13_modules_branch0_modules_bn_parameters_weight_
+        l_self_modules_features_modules_13_modules_branch0_modules_bn_parameters_bias_ = L_self_modules_features_modules_13_modules_branch0_modules_bn_parameters_bias_
+        l_self_modules_features_modules_13_modules_branch1_modules_0_modules_conv_parameters_weight_ = L_self_modules_features_modules_13_modules_branch1_modules_0_modules_conv_parameters_weight_
+        l_self_modules_features_modules_13_modules_branch1_modules_0_modules_bn_buffers_running_mean_ = L_self_modules_features_modules_13_modules_branch1_modules_0_modules_bn_buffers_running_mean_
+        l_self_modules_features_modules_13_modules_branch1_modules_0_modules_bn_buffers_running_var_ = L_self_modules_features_modules_13_modules_branch1_modules_0_modules_bn_buffers_running_var_
+        l_self_modules_features_modules_13_modules_branch1_modules_0_modules_bn_parameters_weight_ = L_self_modules_features_modules_13_modules_branch1_modules_0_modules_bn_parameters_weight_
+        l_self_modules_features_modules_13_modules_branch1_modules_0_modules_bn_parameters_bias_ = L_self_modules_features_modules_13_modules_branch1_modules_0_modules_bn_parameters_bias_
+        l_self_modules_features_modules_13_modules_branch1_modules_1_modules_conv_parameters_weight_ = L_self_modules_features_modules_13_modules_branch1_modules_1_modules_conv_parameters_weight_
+        l_self_modules_features_modules_13_modules_branch1_modules_1_modules_bn_buffers_running_mean_ = L_self_modules_features_modules_13_modules_branch1_modules_1_modules_bn_buffers_running_mean_
+        l_self_modules_features_modules_13_modules_branch1_modules_1_modules_bn_buffers_running_var_ = L_self_modules_features_modules_13_modules_branch1_modules_1_modules_bn_buffers_running_var_
+        l_self_modules_features_modules_13_modules_branch1_modules_1_modules_bn_parameters_weight_ = L_self_modules_features_modules_13_modules_branch1_modules_1_modules_bn_parameters_weight_
+        l_self_modules_features_modules_13_modules_branch1_modules_1_modules_bn_parameters_bias_ = L_self_modules_features_modules_13_modules_branch1_modules_1_modules_bn_parameters_bias_
+        l_self_modules_features_modules_13_modules_branch1_modules_2_modules_conv_parameters_weight_ = L_self_modules_features_modules_13_modules_branch1_modules_2_modules_conv_parameters_weight_
+        l_self_modules_features_modules_13_modules_branch1_modules_2_modules_bn_buffers_running_mean_ = L_self_modules_features_modules_13_modules_branch1_modules_2_modules_bn_buffers_running_mean_
+        l_self_modules_features_modules_13_modules_branch1_modules_2_modules_bn_buffers_running_var_ = L_self_modules_features_modules_13_modules_branch1_modules_2_modules_bn_buffers_running_var_
+        l_self_modules_features_modules_13_modules_branch1_modules_2_modules_bn_parameters_weight_ = L_self_modules_features_modules_13_modules_branch1_modules_2_modules_bn_parameters_weight_
+        l_self_modules_features_modules_13_modules_branch1_modules_2_modules_bn_parameters_bias_ = L_self_modules_features_modules_13_modules_branch1_modules_2_modules_bn_parameters_bias_
+        l_self_modules_features_modules_13_modules_branch2_modules_0_modules_conv_parameters_weight_ = L_self_modules_features_modules_13_modules_branch2_modules_0_modules_conv_parameters_weight_
+        l_self_modules_features_modules_13_modules_branch2_modules_0_modules_bn_buffers_running_mean_ = L_self_modules_features_modules_13_modules_branch2_modules_0_modules_bn_buffers_running_mean_
+        l_self_modules_features_modules_13_modules_branch2_modules_0_modules_bn_buffers_running_var_ = L_self_modules_features_modules_13_modules_branch2_modules_0_modules_bn_buffers_running_var_
+        l_self_modules_features_modules_13_modules_branch2_modules_0_modules_bn_parameters_weight_ = L_self_modules_features_modules_13_modules_branch2_modules_0_modules_bn_parameters_weight_
+        l_self_modules_features_modules_13_modules_branch2_modules_0_modules_bn_parameters_bias_ = L_self_modules_features_modules_13_modules_branch2_modules_0_modules_bn_parameters_bias_
+        l_self_modules_features_modules_13_modules_branch2_modules_1_modules_conv_parameters_weight_ = L_self_modules_features_modules_13_modules_branch2_modules_1_modules_conv_parameters_weight_
+        l_self_modules_features_modules_13_modules_branch2_modules_1_modules_bn_buffers_running_mean_ = L_self_modules_features_modules_13_modules_branch2_modules_1_modules_bn_buffers_running_mean_
+        l_self_modules_features_modules_13_modules_branch2_modules_1_modules_bn_buffers_running_var_ = L_self_modules_features_modules_13_modules_branch2_modules_1_modules_bn_buffers_running_var_
+        l_self_modules_features_modules_13_modules_branch2_modules_1_modules_bn_parameters_weight_ = L_self_modules_features_modules_13_modules_branch2_modules_1_modules_bn_parameters_weight_
+        l_self_modules_features_modules_13_modules_branch2_modules_1_modules_bn_parameters_bias_ = L_self_modules_features_modules_13_modules_branch2_modules_1_modules_bn_parameters_bias_
+        l_self_modules_features_modules_13_modules_branch2_modules_2_modules_conv_parameters_weight_ = L_self_modules_features_modules_13_modules_branch2_modules_2_modules_conv_parameters_weight_
+        l_self_modules_features_modules_13_modules_branch2_modules_2_modules_bn_buffers_running_mean_ = L_self_modules_features_modules_13_modules_branch2_modules_2_modules_bn_buffers_running_mean_
+        l_self_modules_features_modules_13_modules_branch2_modules_2_modules_bn_buffers_running_var_ = L_self_modules_features_modules_13_modules_branch2_modules_2_modules_bn_buffers_running_var_
+        l_self_modules_features_modules_13_modules_branch2_modules_2_modules_bn_parameters_weight_ = L_self_modules_features_modules_13_modules_branch2_modules_2_modules_bn_parameters_weight_
+        l_self_modules_features_modules_13_modules_branch2_modules_2_modules_bn_parameters_bias_ = L_self_modules_features_modules_13_modules_branch2_modules_2_modules_bn_parameters_bias_
+        l_self_modules_features_modules_13_modules_branch2_modules_3_modules_conv_parameters_weight_ = L_self_modules_features_modules_13_modules_branch2_modules_3_modules_conv_parameters_weight_
+        l_self_modules_features_modules_13_modules_branch2_modules_3_modules_bn_buffers_running_mean_ = L_self_modules_features_modules_13_modules_branch2_modules_3_modules_bn_buffers_running_mean_
+        l_self_modules_features_modules_13_modules_branch2_modules_3_modules_bn_buffers_running_var_ = L_self_modules_features_modules_13_modules_branch2_modules_3_modules_bn_buffers_running_var_
+        l_self_modules_features_modules_13_modules_branch2_modules_3_modules_bn_parameters_weight_ = L_self_modules_features_modules_13_modules_branch2_modules_3_modules_bn_parameters_weight_
+        l_self_modules_features_modules_13_modules_branch2_modules_3_modules_bn_parameters_bias_ = L_self_modules_features_modules_13_modules_branch2_modules_3_modules_bn_parameters_bias_
+        l_self_modules_features_modules_13_modules_branch2_modules_4_modules_conv_parameters_weight_ = L_self_modules_features_modules_13_modules_branch2_modules_4_modules_conv_parameters_weight_
+        l_self_modules_features_modules_13_modules_branch2_modules_4_modules_bn_buffers_running_mean_ = L_self_modules_features_modules_13_modules_branch2_modules_4_modules_bn_buffers_running_mean_
+        l_self_modules_features_modules_13_modules_branch2_modules_4_modules_bn_buffers_running_var_ = L_self_modules_features_modules_13_modules_branch2_modules_4_modules_bn_buffers_running_var_
+        l_self_modules_features_modules_13_modules_branch2_modules_4_modules_bn_parameters_weight_ = L_self_modules_features_modules_13_modules_branch2_modules_4_modules_bn_parameters_weight_
+        l_self_modules_features_modules_13_modules_branch2_modules_4_modules_bn_parameters_bias_ = L_self_modules_features_modules_13_modules_branch2_modules_4_modules_bn_parameters_bias_
+        l_self_modules_features_modules_13_modules_branch3_modules_1_modules_conv_parameters_weight_ = L_self_modules_features_modules_13_modules_branch3_modules_1_modules_conv_parameters_weight_
+        l_self_modules_features_modules_13_modules_branch3_modules_1_modules_bn_buffers_running_mean_ = L_self_modules_features_modules_13_modules_branch3_modules_1_modules_bn_buffers_running_mean_
+        l_self_modules_features_modules_13_modules_branch3_modules_1_modules_bn_buffers_running_var_ = L_self_modules_features_modules_13_modules_branch3_modules_1_modules_bn_buffers_running_var_
+        l_self_modules_features_modules_13_modules_branch3_modules_1_modules_bn_parameters_weight_ = L_self_modules_features_modules_13_modules_branch3_modules_1_modules_bn_parameters_weight_
+        l_self_modules_features_modules_13_modules_branch3_modules_1_modules_bn_parameters_bias_ = L_self_modules_features_modules_13_modules_branch3_modules_1_modules_bn_parameters_bias_
+        l_self_modules_features_modules_14_modules_branch0_modules_conv_parameters_weight_ = L_self_modules_features_modules_14_modules_branch0_modules_conv_parameters_weight_
+        l_self_modules_features_modules_14_modules_branch0_modules_bn_buffers_running_mean_ = L_self_modules_features_modules_14_modules_branch0_modules_bn_buffers_running_mean_
+        l_self_modules_features_modules_14_modules_branch0_modules_bn_buffers_running_var_ = L_self_modules_features_modules_14_modules_branch0_modules_bn_buffers_running_var_
+        l_self_modules_features_modules_14_modules_branch0_modules_bn_parameters_weight_ = L_self_modules_features_modules_14_modules_branch0_modules_bn_parameters_weight_
+        l_self_modules_features_modules_14_modules_branch0_modules_bn_parameters_bias_ = L_self_modules_features_modules_14_modules_branch0_modules_bn_parameters_bias_
+        l_self_modules_features_modules_14_modules_branch1_modules_0_modules_conv_parameters_weight_ = L_self_modules_features_modules_14_modules_branch1_modules_0_modules_conv_parameters_weight_
+        l_self_modules_features_modules_14_modules_branch1_modules_0_modules_bn_buffers_running_mean_ = L_self_modules_features_modules_14_modules_branch1_modules_0_modules_bn_buffers_running_mean_
+        l_self_modules_features_modules_14_modules_branch1_modules_0_modules_bn_buffers_running_var_ = L_self_modules_features_modules_14_modules_branch1_modules_0_modules_bn_buffers_running_var_
+        l_self_modules_features_modules_14_modules_branch1_modules_0_modules_bn_parameters_weight_ = L_self_modules_features_modules_14_modules_branch1_modules_0_modules_bn_parameters_weight_
+        l_self_modules_features_modules_14_modules_branch1_modules_0_modules_bn_parameters_bias_ = L_self_modules_features_modules_14_modules_branch1_modules_0_modules_bn_parameters_bias_
+        l_self_modules_features_modules_14_modules_branch1_modules_1_modules_conv_parameters_weight_ = L_self_modules_features_modules_14_modules_branch1_modules_1_modules_conv_parameters_weight_
+        l_self_modules_features_modules_14_modules_branch1_modules_1_modules_bn_buffers_running_mean_ = L_self_modules_features_modules_14_modules_branch1_modules_1_modules_bn_buffers_running_mean_
+        l_self_modules_features_modules_14_modules_branch1_modules_1_modules_bn_buffers_running_var_ = L_self_modules_features_modules_14_modules_branch1_modules_1_modules_bn_buffers_running_var_
+        l_self_modules_features_modules_14_modules_branch1_modules_1_modules_bn_parameters_weight_ = L_self_modules_features_modules_14_modules_branch1_modules_1_modules_bn_parameters_weight_
+        l_self_modules_features_modules_14_modules_branch1_modules_1_modules_bn_parameters_bias_ = L_self_modules_features_modules_14_modules_branch1_modules_1_modules_bn_parameters_bias_
+        l_self_modules_features_modules_14_modules_branch1_modules_2_modules_conv_parameters_weight_ = L_self_modules_features_modules_14_modules_branch1_modules_2_modules_conv_parameters_weight_
+        l_self_modules_features_modules_14_modules_branch1_modules_2_modules_bn_buffers_running_mean_ = L_self_modules_features_modules_14_modules_branch1_modules_2_modules_bn_buffers_running_mean_
+        l_self_modules_features_modules_14_modules_branch1_modules_2_modules_bn_buffers_running_var_ = L_self_modules_features_modules_14_modules_branch1_modules_2_modules_bn_buffers_running_var_
+        l_self_modules_features_modules_14_modules_branch1_modules_2_modules_bn_parameters_weight_ = L_self_modules_features_modules_14_modules_branch1_modules_2_modules_bn_parameters_weight_
+        l_self_modules_features_modules_14_modules_branch1_modules_2_modules_bn_parameters_bias_ = L_self_modules_features_modules_14_modules_branch1_modules_2_modules_bn_parameters_bias_
+        l_self_modules_features_modules_14_modules_branch2_modules_0_modules_conv_parameters_weight_ = L_self_modules_features_modules_14_modules_branch2_modules_0_modules_conv_parameters_weight_
+        l_self_modules_features_modules_14_modules_branch2_modules_0_modules_bn_buffers_running_mean_ = L_self_modules_features_modules_14_modules_branch2_modules_0_modules_bn_buffers_running_mean_
+        l_self_modules_features_modules_14_modules_branch2_modules_0_modules_bn_buffers_running_var_ = L_self_modules_features_modules_14_modules_branch2_modules_0_modules_bn_buffers_running_var_
+        l_self_modules_features_modules_14_modules_branch2_modules_0_modules_bn_parameters_weight_ = L_self_modules_features_modules_14_modules_branch2_modules_0_modules_bn_parameters_weight_
+        l_self_modules_features_modules_14_modules_branch2_modules_0_modules_bn_parameters_bias_ = L_self_modules_features_modules_14_modules_branch2_modules_0_modules_bn_parameters_bias_
+        l_self_modules_features_modules_14_modules_branch2_modules_1_modules_conv_parameters_weight_ = L_self_modules_features_modules_14_modules_branch2_modules_1_modules_conv_parameters_weight_
+        l_self_modules_features_modules_14_modules_branch2_modules_1_modules_bn_buffers_running_mean_ = L_self_modules_features_modules_14_modules_branch2_modules_1_modules_bn_buffers_running_mean_
+        l_self_modules_features_modules_14_modules_branch2_modules_1_modules_bn_buffers_running_var_ = L_self_modules_features_modules_14_modules_branch2_modules_1_modules_bn_buffers_running_var_
+        l_self_modules_features_modules_14_modules_branch2_modules_1_modules_bn_parameters_weight_ = L_self_modules_features_modules_14_modules_branch2_modules_1_modules_bn_parameters_weight_
+        l_self_modules_features_modules_14_modules_branch2_modules_1_modules_bn_parameters_bias_ = L_self_modules_features_modules_14_modules_branch2_modules_1_modules_bn_parameters_bias_
+        l_self_modules_features_modules_14_modules_branch2_modules_2_modules_conv_parameters_weight_ = L_self_modules_features_modules_14_modules_branch2_modules_2_modules_conv_parameters_weight_
+        l_self_modules_features_modules_14_modules_branch2_modules_2_modules_bn_buffers_running_mean_ = L_self_modules_features_modules_14_modules_branch2_modules_2_modules_bn_buffers_running_mean_
+        l_self_modules_features_modules_14_modules_branch2_modules_2_modules_bn_buffers_running_var_ = L_self_modules_features_modules_14_modules_branch2_modules_2_modules_bn_buffers_running_var_
+        l_self_modules_features_modules_14_modules_branch2_modules_2_modules_bn_parameters_weight_ = L_self_modules_features_modules_14_modules_branch2_modules_2_modules_bn_parameters_weight_
+        l_self_modules_features_modules_14_modules_branch2_modules_2_modules_bn_parameters_bias_ = L_self_modules_features_modules_14_modules_branch2_modules_2_modules_bn_parameters_bias_
+        l_self_modules_features_modules_14_modules_branch2_modules_3_modules_conv_parameters_weight_ = L_self_modules_features_modules_14_modules_branch2_modules_3_modules_conv_parameters_weight_
+        l_self_modules_features_modules_14_modules_branch2_modules_3_modules_bn_buffers_running_mean_ = L_self_modules_features_modules_14_modules_branch2_modules_3_modules_bn_buffers_running_mean_
+        l_self_modules_features_modules_14_modules_branch2_modules_3_modules_bn_buffers_running_var_ = L_self_modules_features_modules_14_modules_branch2_modules_3_modules_bn_buffers_running_var_
+        l_self_modules_features_modules_14_modules_branch2_modules_3_modules_bn_parameters_weight_ = L_self_modules_features_modules_14_modules_branch2_modules_3_modules_bn_parameters_weight_
+        l_self_modules_features_modules_14_modules_branch2_modules_3_modules_bn_parameters_bias_ = L_self_modules_features_modules_14_modules_branch2_modules_3_modules_bn_parameters_bias_
+        l_self_modules_features_modules_14_modules_branch2_modules_4_modules_conv_parameters_weight_ = L_self_modules_features_modules_14_modules_branch2_modules_4_modules_conv_parameters_weight_
+        l_self_modules_features_modules_14_modules_branch2_modules_4_modules_bn_buffers_running_mean_ = L_self_modules_features_modules_14_modules_branch2_modules_4_modules_bn_buffers_running_mean_
+        l_self_modules_features_modules_14_modules_branch2_modules_4_modules_bn_buffers_running_var_ = L_self_modules_features_modules_14_modules_branch2_modules_4_modules_bn_buffers_running_var_
+        l_self_modules_features_modules_14_modules_branch2_modules_4_modules_bn_parameters_weight_ = L_self_modules_features_modules_14_modules_branch2_modules_4_modules_bn_parameters_weight_
+        l_self_modules_features_modules_14_modules_branch2_modules_4_modules_bn_parameters_bias_ = L_self_modules_features_modules_14_modules_branch2_modules_4_modules_bn_parameters_bias_
+        l_self_modules_features_modules_14_modules_branch3_modules_1_modules_conv_parameters_weight_ = L_self_modules_features_modules_14_modules_branch3_modules_1_modules_conv_parameters_weight_
+        l_self_modules_features_modules_14_modules_branch3_modules_1_modules_bn_buffers_running_mean_ = L_self_modules_features_modules_14_modules_branch3_modules_1_modules_bn_buffers_running_mean_
+        l_self_modules_features_modules_14_modules_branch3_modules_1_modules_bn_buffers_running_var_ = L_self_modules_features_modules_14_modules_branch3_modules_1_modules_bn_buffers_running_var_
+        l_self_modules_features_modules_14_modules_branch3_modules_1_modules_bn_parameters_weight_ = L_self_modules_features_modules_14_modules_branch3_modules_1_modules_bn_parameters_weight_
+        l_self_modules_features_modules_14_modules_branch3_modules_1_modules_bn_parameters_bias_ = L_self_modules_features_modules_14_modules_branch3_modules_1_modules_bn_parameters_bias_
+        l_self_modules_features_modules_15_modules_branch0_modules_conv_parameters_weight_ = L_self_modules_features_modules_15_modules_branch0_modules_conv_parameters_weight_
+        l_self_modules_features_modules_15_modules_branch0_modules_bn_buffers_running_mean_ = L_self_modules_features_modules_15_modules_branch0_modules_bn_buffers_running_mean_
+        l_self_modules_features_modules_15_modules_branch0_modules_bn_buffers_running_var_ = L_self_modules_features_modules_15_modules_branch0_modules_bn_buffers_running_var_
+        l_self_modules_features_modules_15_modules_branch0_modules_bn_parameters_weight_ = L_self_modules_features_modules_15_modules_branch0_modules_bn_parameters_weight_
+        l_self_modules_features_modules_15_modules_branch0_modules_bn_parameters_bias_ = L_self_modules_features_modules_15_modules_branch0_modules_bn_parameters_bias_
+        l_self_modules_features_modules_15_modules_branch1_modules_0_modules_conv_parameters_weight_ = L_self_modules_features_modules_15_modules_branch1_modules_0_modules_conv_parameters_weight_
+        l_self_modules_features_modules_15_modules_branch1_modules_0_modules_bn_buffers_running_mean_ = L_self_modules_features_modules_15_modules_branch1_modules_0_modules_bn_buffers_running_mean_
+        l_self_modules_features_modules_15_modules_branch1_modules_0_modules_bn_buffers_running_var_ = L_self_modules_features_modules_15_modules_branch1_modules_0_modules_bn_buffers_running_var_
+        l_self_modules_features_modules_15_modules_branch1_modules_0_modules_bn_parameters_weight_ = L_self_modules_features_modules_15_modules_branch1_modules_0_modules_bn_parameters_weight_
+        l_self_modules_features_modules_15_modules_branch1_modules_0_modules_bn_parameters_bias_ = L_self_modules_features_modules_15_modules_branch1_modules_0_modules_bn_parameters_bias_
+        l_self_modules_features_modules_15_modules_branch1_modules_1_modules_conv_parameters_weight_ = L_self_modules_features_modules_15_modules_branch1_modules_1_modules_conv_parameters_weight_
+        l_self_modules_features_modules_15_modules_branch1_modules_1_modules_bn_buffers_running_mean_ = L_self_modules_features_modules_15_modules_branch1_modules_1_modules_bn_buffers_running_mean_
+        l_self_modules_features_modules_15_modules_branch1_modules_1_modules_bn_buffers_running_var_ = L_self_modules_features_modules_15_modules_branch1_modules_1_modules_bn_buffers_running_var_
+        l_self_modules_features_modules_15_modules_branch1_modules_1_modules_bn_parameters_weight_ = L_self_modules_features_modules_15_modules_branch1_modules_1_modules_bn_parameters_weight_
+        l_self_modules_features_modules_15_modules_branch1_modules_1_modules_bn_parameters_bias_ = L_self_modules_features_modules_15_modules_branch1_modules_1_modules_bn_parameters_bias_
+        l_self_modules_features_modules_15_modules_branch1_modules_2_modules_conv_parameters_weight_ = L_self_modules_features_modules_15_modules_branch1_modules_2_modules_conv_parameters_weight_
+        l_self_modules_features_modules_15_modules_branch1_modules_2_modules_bn_buffers_running_mean_ = L_self_modules_features_modules_15_modules_branch1_modules_2_modules_bn_buffers_running_mean_
+        l_self_modules_features_modules_15_modules_branch1_modules_2_modules_bn_buffers_running_var_ = L_self_modules_features_modules_15_modules_branch1_modules_2_modules_bn_buffers_running_var_
+        l_self_modules_features_modules_15_modules_branch1_modules_2_modules_bn_parameters_weight_ = L_self_modules_features_modules_15_modules_branch1_modules_2_modules_bn_parameters_weight_
+        l_self_modules_features_modules_15_modules_branch1_modules_2_modules_bn_parameters_bias_ = L_self_modules_features_modules_15_modules_branch1_modules_2_modules_bn_parameters_bias_
+        l_self_modules_features_modules_15_modules_branch2_modules_0_modules_conv_parameters_weight_ = L_self_modules_features_modules_15_modules_branch2_modules_0_modules_conv_parameters_weight_
+        l_self_modules_features_modules_15_modules_branch2_modules_0_modules_bn_buffers_running_mean_ = L_self_modules_features_modules_15_modules_branch2_modules_0_modules_bn_buffers_running_mean_
+        l_self_modules_features_modules_15_modules_branch2_modules_0_modules_bn_buffers_running_var_ = L_self_modules_features_modules_15_modules_branch2_modules_0_modules_bn_buffers_running_var_
+        l_self_modules_features_modules_15_modules_branch2_modules_0_modules_bn_parameters_weight_ = L_self_modules_features_modules_15_modules_branch2_modules_0_modules_bn_parameters_weight_
+        l_self_modules_features_modules_15_modules_branch2_modules_0_modules_bn_parameters_bias_ = L_self_modules_features_modules_15_modules_branch2_modules_0_modules_bn_parameters_bias_
+        l_self_modules_features_modules_15_modules_branch2_modules_1_modules_conv_parameters_weight_ = L_self_modules_features_modules_15_modules_branch2_modules_1_modules_conv_parameters_weight_
+        l_self_modules_features_modules_15_modules_branch2_modules_1_modules_bn_buffers_running_mean_ = L_self_modules_features_modules_15_modules_branch2_modules_1_modules_bn_buffers_running_mean_
+        l_self_modules_features_modules_15_modules_branch2_modules_1_modules_bn_buffers_running_var_ = L_self_modules_features_modules_15_modules_branch2_modules_1_modules_bn_buffers_running_var_
+        l_self_modules_features_modules_15_modules_branch2_modules_1_modules_bn_parameters_weight_ = L_self_modules_features_modules_15_modules_branch2_modules_1_modules_bn_parameters_weight_
+        l_self_modules_features_modules_15_modules_branch2_modules_1_modules_bn_parameters_bias_ = L_self_modules_features_modules_15_modules_branch2_modules_1_modules_bn_parameters_bias_
+        l_self_modules_features_modules_15_modules_branch2_modules_2_modules_conv_parameters_weight_ = L_self_modules_features_modules_15_modules_branch2_modules_2_modules_conv_parameters_weight_
+        l_self_modules_features_modules_15_modules_branch2_modules_2_modules_bn_buffers_running_mean_ = L_self_modules_features_modules_15_modules_branch2_modules_2_modules_bn_buffers_running_mean_
+        l_self_modules_features_modules_15_modules_branch2_modules_2_modules_bn_buffers_running_var_ = L_self_modules_features_modules_15_modules_branch2_modules_2_modules_bn_buffers_running_var_
+        l_self_modules_features_modules_15_modules_branch2_modules_2_modules_bn_parameters_weight_ = L_self_modules_features_modules_15_modules_branch2_modules_2_modules_bn_parameters_weight_
+        l_self_modules_features_modules_15_modules_branch2_modules_2_modules_bn_parameters_bias_ = L_self_modules_features_modules_15_modules_branch2_modules_2_modules_bn_parameters_bias_
+        l_self_modules_features_modules_15_modules_branch2_modules_3_modules_conv_parameters_weight_ = L_self_modules_features_modules_15_modules_branch2_modules_3_modules_conv_parameters_weight_
+        l_self_modules_features_modules_15_modules_branch2_modules_3_modules_bn_buffers_running_mean_ = L_self_modules_features_modules_15_modules_branch2_modules_3_modules_bn_buffers_running_mean_
+        l_self_modules_features_modules_15_modules_branch2_modules_3_modules_bn_buffers_running_var_ = L_self_modules_features_modules_15_modules_branch2_modules_3_modules_bn_buffers_running_var_
+        l_self_modules_features_modules_15_modules_branch2_modules_3_modules_bn_parameters_weight_ = L_self_modules_features_modules_15_modules_branch2_modules_3_modules_bn_parameters_weight_
+        l_self_modules_features_modules_15_modules_branch2_modules_3_modules_bn_parameters_bias_ = L_self_modules_features_modules_15_modules_branch2_modules_3_modules_bn_parameters_bias_
+        l_self_modules_features_modules_15_modules_branch2_modules_4_modules_conv_parameters_weight_ = L_self_modules_features_modules_15_modules_branch2_modules_4_modules_conv_parameters_weight_
+        l_self_modules_features_modules_15_modules_branch2_modules_4_modules_bn_buffers_running_mean_ = L_self_modules_features_modules_15_modules_branch2_modules_4_modules_bn_buffers_running_mean_
+        l_self_modules_features_modules_15_modules_branch2_modules_4_modules_bn_buffers_running_var_ = L_self_modules_features_modules_15_modules_branch2_modules_4_modules_bn_buffers_running_var_
+        l_self_modules_features_modules_15_modules_branch2_modules_4_modules_bn_parameters_weight_ = L_self_modules_features_modules_15_modules_branch2_modules_4_modules_bn_parameters_weight_
+        l_self_modules_features_modules_15_modules_branch2_modules_4_modules_bn_parameters_bias_ = L_self_modules_features_modules_15_modules_branch2_modules_4_modules_bn_parameters_bias_
+        l_self_modules_features_modules_15_modules_branch3_modules_1_modules_conv_parameters_weight_ = L_self_modules_features_modules_15_modules_branch3_modules_1_modules_conv_parameters_weight_
+        l_self_modules_features_modules_15_modules_branch3_modules_1_modules_bn_buffers_running_mean_ = L_self_modules_features_modules_15_modules_branch3_modules_1_modules_bn_buffers_running_mean_
+        l_self_modules_features_modules_15_modules_branch3_modules_1_modules_bn_buffers_running_var_ = L_self_modules_features_modules_15_modules_branch3_modules_1_modules_bn_buffers_running_var_
+        l_self_modules_features_modules_15_modules_branch3_modules_1_modules_bn_parameters_weight_ = L_self_modules_features_modules_15_modules_branch3_modules_1_modules_bn_parameters_weight_
+        l_self_modules_features_modules_15_modules_branch3_modules_1_modules_bn_parameters_bias_ = L_self_modules_features_modules_15_modules_branch3_modules_1_modules_bn_parameters_bias_
+        l_self_modules_features_modules_16_modules_branch0_modules_conv_parameters_weight_ = L_self_modules_features_modules_16_modules_branch0_modules_conv_parameters_weight_
+        l_self_modules_features_modules_16_modules_branch0_modules_bn_buffers_running_mean_ = L_self_modules_features_modules_16_modules_branch0_modules_bn_buffers_running_mean_
+        l_self_modules_features_modules_16_modules_branch0_modules_bn_buffers_running_var_ = L_self_modules_features_modules_16_modules_branch0_modules_bn_buffers_running_var_
+        l_self_modules_features_modules_16_modules_branch0_modules_bn_parameters_weight_ = L_self_modules_features_modules_16_modules_branch0_modules_bn_parameters_weight_
+        l_self_modules_features_modules_16_modules_branch0_modules_bn_parameters_bias_ = L_self_modules_features_modules_16_modules_branch0_modules_bn_parameters_bias_
+        l_self_modules_features_modules_16_modules_branch1_modules_0_modules_conv_parameters_weight_ = L_self_modules_features_modules_16_modules_branch1_modules_0_modules_conv_parameters_weight_
+        l_self_modules_features_modules_16_modules_branch1_modules_0_modules_bn_buffers_running_mean_ = L_self_modules_features_modules_16_modules_branch1_modules_0_modules_bn_buffers_running_mean_
+        l_self_modules_features_modules_16_modules_branch1_modules_0_modules_bn_buffers_running_var_ = L_self_modules_features_modules_16_modules_branch1_modules_0_modules_bn_buffers_running_var_
+        l_self_modules_features_modules_16_modules_branch1_modules_0_modules_bn_parameters_weight_ = L_self_modules_features_modules_16_modules_branch1_modules_0_modules_bn_parameters_weight_
+        l_self_modules_features_modules_16_modules_branch1_modules_0_modules_bn_parameters_bias_ = L_self_modules_features_modules_16_modules_branch1_modules_0_modules_bn_parameters_bias_
+        l_self_modules_features_modules_16_modules_branch1_modules_1_modules_conv_parameters_weight_ = L_self_modules_features_modules_16_modules_branch1_modules_1_modules_conv_parameters_weight_
+        l_self_modules_features_modules_16_modules_branch1_modules_1_modules_bn_buffers_running_mean_ = L_self_modules_features_modules_16_modules_branch1_modules_1_modules_bn_buffers_running_mean_
+        l_self_modules_features_modules_16_modules_branch1_modules_1_modules_bn_buffers_running_var_ = L_self_modules_features_modules_16_modules_branch1_modules_1_modules_bn_buffers_running_var_
+        l_self_modules_features_modules_16_modules_branch1_modules_1_modules_bn_parameters_weight_ = L_self_modules_features_modules_16_modules_branch1_modules_1_modules_bn_parameters_weight_
+        l_self_modules_features_modules_16_modules_branch1_modules_1_modules_bn_parameters_bias_ = L_self_modules_features_modules_16_modules_branch1_modules_1_modules_bn_parameters_bias_
+        l_self_modules_features_modules_16_modules_branch1_modules_2_modules_conv_parameters_weight_ = L_self_modules_features_modules_16_modules_branch1_modules_2_modules_conv_parameters_weight_
+        l_self_modules_features_modules_16_modules_branch1_modules_2_modules_bn_buffers_running_mean_ = L_self_modules_features_modules_16_modules_branch1_modules_2_modules_bn_buffers_running_mean_
+        l_self_modules_features_modules_16_modules_branch1_modules_2_modules_bn_buffers_running_var_ = L_self_modules_features_modules_16_modules_branch1_modules_2_modules_bn_buffers_running_var_
+        l_self_modules_features_modules_16_modules_branch1_modules_2_modules_bn_parameters_weight_ = L_self_modules_features_modules_16_modules_branch1_modules_2_modules_bn_parameters_weight_
+        l_self_modules_features_modules_16_modules_branch1_modules_2_modules_bn_parameters_bias_ = L_self_modules_features_modules_16_modules_branch1_modules_2_modules_bn_parameters_bias_
+        l_self_modules_features_modules_16_modules_branch2_modules_0_modules_conv_parameters_weight_ = L_self_modules_features_modules_16_modules_branch2_modules_0_modules_conv_parameters_weight_
+        l_self_modules_features_modules_16_modules_branch2_modules_0_modules_bn_buffers_running_mean_ = L_self_modules_features_modules_16_modules_branch2_modules_0_modules_bn_buffers_running_mean_
+        l_self_modules_features_modules_16_modules_branch2_modules_0_modules_bn_buffers_running_var_ = L_self_modules_features_modules_16_modules_branch2_modules_0_modules_bn_buffers_running_var_
+        l_self_modules_features_modules_16_modules_branch2_modules_0_modules_bn_parameters_weight_ = L_self_modules_features_modules_16_modules_branch2_modules_0_modules_bn_parameters_weight_
+        l_self_modules_features_modules_16_modules_branch2_modules_0_modules_bn_parameters_bias_ = L_self_modules_features_modules_16_modules_branch2_modules_0_modules_bn_parameters_bias_
+        l_self_modules_features_modules_16_modules_branch2_modules_1_modules_conv_parameters_weight_ = L_self_modules_features_modules_16_modules_branch2_modules_1_modules_conv_parameters_weight_
+        l_self_modules_features_modules_16_modules_branch2_modules_1_modules_bn_buffers_running_mean_ = L_self_modules_features_modules_16_modules_branch2_modules_1_modules_bn_buffers_running_mean_
+        l_self_modules_features_modules_16_modules_branch2_modules_1_modules_bn_buffers_running_var_ = L_self_modules_features_modules_16_modules_branch2_modules_1_modules_bn_buffers_running_var_
+        l_self_modules_features_modules_16_modules_branch2_modules_1_modules_bn_parameters_weight_ = L_self_modules_features_modules_16_modules_branch2_modules_1_modules_bn_parameters_weight_
+        l_self_modules_features_modules_16_modules_branch2_modules_1_modules_bn_parameters_bias_ = L_self_modules_features_modules_16_modules_branch2_modules_1_modules_bn_parameters_bias_
+        l_self_modules_features_modules_16_modules_branch2_modules_2_modules_conv_parameters_weight_ = L_self_modules_features_modules_16_modules_branch2_modules_2_modules_conv_parameters_weight_
+        l_self_modules_features_modules_16_modules_branch2_modules_2_modules_bn_buffers_running_mean_ = L_self_modules_features_modules_16_modules_branch2_modules_2_modules_bn_buffers_running_mean_
+        l_self_modules_features_modules_16_modules_branch2_modules_2_modules_bn_buffers_running_var_ = L_self_modules_features_modules_16_modules_branch2_modules_2_modules_bn_buffers_running_var_
+        l_self_modules_features_modules_16_modules_branch2_modules_2_modules_bn_parameters_weight_ = L_self_modules_features_modules_16_modules_branch2_modules_2_modules_bn_parameters_weight_
+        l_self_modules_features_modules_16_modules_branch2_modules_2_modules_bn_parameters_bias_ = L_self_modules_features_modules_16_modules_branch2_modules_2_modules_bn_parameters_bias_
+        l_self_modules_features_modules_16_modules_branch2_modules_3_modules_conv_parameters_weight_ = L_self_modules_features_modules_16_modules_branch2_modules_3_modules_conv_parameters_weight_
+        l_self_modules_features_modules_16_modules_branch2_modules_3_modules_bn_buffers_running_mean_ = L_self_modules_features_modules_16_modules_branch2_modules_3_modules_bn_buffers_running_mean_
+        l_self_modules_features_modules_16_modules_branch2_modules_3_modules_bn_buffers_running_var_ = L_self_modules_features_modules_16_modules_branch2_modules_3_modules_bn_buffers_running_var_
+        l_self_modules_features_modules_16_modules_branch2_modules_3_modules_bn_parameters_weight_ = L_self_modules_features_modules_16_modules_branch2_modules_3_modules_bn_parameters_weight_
+        l_self_modules_features_modules_16_modules_branch2_modules_3_modules_bn_parameters_bias_ = L_self_modules_features_modules_16_modules_branch2_modules_3_modules_bn_parameters_bias_
+        l_self_modules_features_modules_16_modules_branch2_modules_4_modules_conv_parameters_weight_ = L_self_modules_features_modules_16_modules_branch2_modules_4_modules_conv_parameters_weight_
+        l_self_modules_features_modules_16_modules_branch2_modules_4_modules_bn_buffers_running_mean_ = L_self_modules_features_modules_16_modules_branch2_modules_4_modules_bn_buffers_running_mean_
+        l_self_modules_features_modules_16_modules_branch2_modules_4_modules_bn_buffers_running_var_ = L_self_modules_features_modules_16_modules_branch2_modules_4_modules_bn_buffers_running_var_
+        l_self_modules_features_modules_16_modules_branch2_modules_4_modules_bn_parameters_weight_ = L_self_modules_features_modules_16_modules_branch2_modules_4_modules_bn_parameters_weight_
+        l_self_modules_features_modules_16_modules_branch2_modules_4_modules_bn_parameters_bias_ = L_self_modules_features_modules_16_modules_branch2_modules_4_modules_bn_parameters_bias_
+        l_self_modules_features_modules_16_modules_branch3_modules_1_modules_conv_parameters_weight_ = L_self_modules_features_modules_16_modules_branch3_modules_1_modules_conv_parameters_weight_
+        l_self_modules_features_modules_16_modules_branch3_modules_1_modules_bn_buffers_running_mean_ = L_self_modules_features_modules_16_modules_branch3_modules_1_modules_bn_buffers_running_mean_
+        l_self_modules_features_modules_16_modules_branch3_modules_1_modules_bn_buffers_running_var_ = L_self_modules_features_modules_16_modules_branch3_modules_1_modules_bn_buffers_running_var_
+        l_self_modules_features_modules_16_modules_branch3_modules_1_modules_bn_parameters_weight_ = L_self_modules_features_modules_16_modules_branch3_modules_1_modules_bn_parameters_weight_
+        l_self_modules_features_modules_16_modules_branch3_modules_1_modules_bn_parameters_bias_ = L_self_modules_features_modules_16_modules_branch3_modules_1_modules_bn_parameters_bias_
+        l_self_modules_features_modules_17_modules_branch0_modules_conv_parameters_weight_ = L_self_modules_features_modules_17_modules_branch0_modules_conv_parameters_weight_
+        l_self_modules_features_modules_17_modules_branch0_modules_bn_buffers_running_mean_ = L_self_modules_features_modules_17_modules_branch0_modules_bn_buffers_running_mean_
+        l_self_modules_features_modules_17_modules_branch0_modules_bn_buffers_running_var_ = L_self_modules_features_modules_17_modules_branch0_modules_bn_buffers_running_var_
+        l_self_modules_features_modules_17_modules_branch0_modules_bn_parameters_weight_ = L_self_modules_features_modules_17_modules_branch0_modules_bn_parameters_weight_
+        l_self_modules_features_modules_17_modules_branch0_modules_bn_parameters_bias_ = L_self_modules_features_modules_17_modules_branch0_modules_bn_parameters_bias_
+        l_self_modules_features_modules_17_modules_branch1_modules_0_modules_conv_parameters_weight_ = L_self_modules_features_modules_17_modules_branch1_modules_0_modules_conv_parameters_weight_
+        l_self_modules_features_modules_17_modules_branch1_modules_0_modules_bn_buffers_running_mean_ = L_self_modules_features_modules_17_modules_branch1_modules_0_modules_bn_buffers_running_mean_
+        l_self_modules_features_modules_17_modules_branch1_modules_0_modules_bn_buffers_running_var_ = L_self_modules_features_modules_17_modules_branch1_modules_0_modules_bn_buffers_running_var_
+        l_self_modules_features_modules_17_modules_branch1_modules_0_modules_bn_parameters_weight_ = L_self_modules_features_modules_17_modules_branch1_modules_0_modules_bn_parameters_weight_
+        l_self_modules_features_modules_17_modules_branch1_modules_0_modules_bn_parameters_bias_ = L_self_modules_features_modules_17_modules_branch1_modules_0_modules_bn_parameters_bias_
+        l_self_modules_features_modules_17_modules_branch1_modules_1_modules_conv_parameters_weight_ = L_self_modules_features_modules_17_modules_branch1_modules_1_modules_conv_parameters_weight_
+        l_self_modules_features_modules_17_modules_branch1_modules_1_modules_bn_buffers_running_mean_ = L_self_modules_features_modules_17_modules_branch1_modules_1_modules_bn_buffers_running_mean_
+        l_self_modules_features_modules_17_modules_branch1_modules_1_modules_bn_buffers_running_var_ = L_self_modules_features_modules_17_modules_branch1_modules_1_modules_bn_buffers_running_var_
+        l_self_modules_features_modules_17_modules_branch1_modules_1_modules_bn_parameters_weight_ = L_self_modules_features_modules_17_modules_branch1_modules_1_modules_bn_parameters_weight_
+        l_self_modules_features_modules_17_modules_branch1_modules_1_modules_bn_parameters_bias_ = L_self_modules_features_modules_17_modules_branch1_modules_1_modules_bn_parameters_bias_
+        l_self_modules_features_modules_17_modules_branch1_modules_2_modules_conv_parameters_weight_ = L_self_modules_features_modules_17_modules_branch1_modules_2_modules_conv_parameters_weight_
+        l_self_modules_features_modules_17_modules_branch1_modules_2_modules_bn_buffers_running_mean_ = L_self_modules_features_modules_17_modules_branch1_modules_2_modules_bn_buffers_running_mean_
+        l_self_modules_features_modules_17_modules_branch1_modules_2_modules_bn_buffers_running_var_ = L_self_modules_features_modules_17_modules_branch1_modules_2_modules_bn_buffers_running_var_
+        l_self_modules_features_modules_17_modules_branch1_modules_2_modules_bn_parameters_weight_ = L_self_modules_features_modules_17_modules_branch1_modules_2_modules_bn_parameters_weight_
+        l_self_modules_features_modules_17_modules_branch1_modules_2_modules_bn_parameters_bias_ = L_self_modules_features_modules_17_modules_branch1_modules_2_modules_bn_parameters_bias_
+        l_self_modules_features_modules_17_modules_branch2_modules_0_modules_conv_parameters_weight_ = L_self_modules_features_modules_17_modules_branch2_modules_0_modules_conv_parameters_weight_
+        l_self_modules_features_modules_17_modules_branch2_modules_0_modules_bn_buffers_running_mean_ = L_self_modules_features_modules_17_modules_branch2_modules_0_modules_bn_buffers_running_mean_
+        l_self_modules_features_modules_17_modules_branch2_modules_0_modules_bn_buffers_running_var_ = L_self_modules_features_modules_17_modules_branch2_modules_0_modules_bn_buffers_running_var_
+        l_self_modules_features_modules_17_modules_branch2_modules_0_modules_bn_parameters_weight_ = L_self_modules_features_modules_17_modules_branch2_modules_0_modules_bn_parameters_weight_
+        l_self_modules_features_modules_17_modules_branch2_modules_0_modules_bn_parameters_bias_ = L_self_modules_features_modules_17_modules_branch2_modules_0_modules_bn_parameters_bias_
+        l_self_modules_features_modules_17_modules_branch2_modules_1_modules_conv_parameters_weight_ = L_self_modules_features_modules_17_modules_branch2_modules_1_modules_conv_parameters_weight_
+        l_self_modules_features_modules_17_modules_branch2_modules_1_modules_bn_buffers_running_mean_ = L_self_modules_features_modules_17_modules_branch2_modules_1_modules_bn_buffers_running_mean_
+        l_self_modules_features_modules_17_modules_branch2_modules_1_modules_bn_buffers_running_var_ = L_self_modules_features_modules_17_modules_branch2_modules_1_modules_bn_buffers_running_var_
+        l_self_modules_features_modules_17_modules_branch2_modules_1_modules_bn_parameters_weight_ = L_self_modules_features_modules_17_modules_branch2_modules_1_modules_bn_parameters_weight_
+        l_self_modules_features_modules_17_modules_branch2_modules_1_modules_bn_parameters_bias_ = L_self_modules_features_modules_17_modules_branch2_modules_1_modules_bn_parameters_bias_
+        l_self_modules_features_modules_17_modules_branch2_modules_2_modules_conv_parameters_weight_ = L_self_modules_features_modules_17_modules_branch2_modules_2_modules_conv_parameters_weight_
+        l_self_modules_features_modules_17_modules_branch2_modules_2_modules_bn_buffers_running_mean_ = L_self_modules_features_modules_17_modules_branch2_modules_2_modules_bn_buffers_running_mean_
+        l_self_modules_features_modules_17_modules_branch2_modules_2_modules_bn_buffers_running_var_ = L_self_modules_features_modules_17_modules_branch2_modules_2_modules_bn_buffers_running_var_
+        l_self_modules_features_modules_17_modules_branch2_modules_2_modules_bn_parameters_weight_ = L_self_modules_features_modules_17_modules_branch2_modules_2_modules_bn_parameters_weight_
+        l_self_modules_features_modules_17_modules_branch2_modules_2_modules_bn_parameters_bias_ = L_self_modules_features_modules_17_modules_branch2_modules_2_modules_bn_parameters_bias_
+        l_self_modules_features_modules_17_modules_branch2_modules_3_modules_conv_parameters_weight_ = L_self_modules_features_modules_17_modules_branch2_modules_3_modules_conv_parameters_weight_
+        l_self_modules_features_modules_17_modules_branch2_modules_3_modules_bn_buffers_running_mean_ = L_self_modules_features_modules_17_modules_branch2_modules_3_modules_bn_buffers_running_mean_
+        l_self_modules_features_modules_17_modules_branch2_modules_3_modules_bn_buffers_running_var_ = L_self_modules_features_modules_17_modules_branch2_modules_3_modules_bn_buffers_running_var_
+        l_self_modules_features_modules_17_modules_branch2_modules_3_modules_bn_parameters_weight_ = L_self_modules_features_modules_17_modules_branch2_modules_3_modules_bn_parameters_weight_
+        l_self_modules_features_modules_17_modules_branch2_modules_3_modules_bn_parameters_bias_ = L_self_modules_features_modules_17_modules_branch2_modules_3_modules_bn_parameters_bias_
+        l_self_modules_features_modules_17_modules_branch2_modules_4_modules_conv_parameters_weight_ = L_self_modules_features_modules_17_modules_branch2_modules_4_modules_conv_parameters_weight_
+        l_self_modules_features_modules_17_modules_branch2_modules_4_modules_bn_buffers_running_mean_ = L_self_modules_features_modules_17_modules_branch2_modules_4_modules_bn_buffers_running_mean_
+        l_self_modules_features_modules_17_modules_branch2_modules_4_modules_bn_buffers_running_var_ = L_self_modules_features_modules_17_modules_branch2_modules_4_modules_bn_buffers_running_var_
+        l_self_modules_features_modules_17_modules_branch2_modules_4_modules_bn_parameters_weight_ = L_self_modules_features_modules_17_modules_branch2_modules_4_modules_bn_parameters_weight_
+        l_self_modules_features_modules_17_modules_branch2_modules_4_modules_bn_parameters_bias_ = L_self_modules_features_modules_17_modules_branch2_modules_4_modules_bn_parameters_bias_
+        l_self_modules_features_modules_17_modules_branch3_modules_1_modules_conv_parameters_weight_ = L_self_modules_features_modules_17_modules_branch3_modules_1_modules_conv_parameters_weight_
+        l_self_modules_features_modules_17_modules_branch3_modules_1_modules_bn_buffers_running_mean_ = L_self_modules_features_modules_17_modules_branch3_modules_1_modules_bn_buffers_running_mean_
+        l_self_modules_features_modules_17_modules_branch3_modules_1_modules_bn_buffers_running_var_ = L_self_modules_features_modules_17_modules_branch3_modules_1_modules_bn_buffers_running_var_
+        l_self_modules_features_modules_17_modules_branch3_modules_1_modules_bn_parameters_weight_ = L_self_modules_features_modules_17_modules_branch3_modules_1_modules_bn_parameters_weight_
+        l_self_modules_features_modules_17_modules_branch3_modules_1_modules_bn_parameters_bias_ = L_self_modules_features_modules_17_modules_branch3_modules_1_modules_bn_parameters_bias_
+        l_self_modules_features_modules_18_modules_branch0_modules_0_modules_conv_parameters_weight_ = L_self_modules_features_modules_18_modules_branch0_modules_0_modules_conv_parameters_weight_
+        l_self_modules_features_modules_18_modules_branch0_modules_0_modules_bn_buffers_running_mean_ = L_self_modules_features_modules_18_modules_branch0_modules_0_modules_bn_buffers_running_mean_
+        l_self_modules_features_modules_18_modules_branch0_modules_0_modules_bn_buffers_running_var_ = L_self_modules_features_modules_18_modules_branch0_modules_0_modules_bn_buffers_running_var_
+        l_self_modules_features_modules_18_modules_branch0_modules_0_modules_bn_parameters_weight_ = L_self_modules_features_modules_18_modules_branch0_modules_0_modules_bn_parameters_weight_
+        l_self_modules_features_modules_18_modules_branch0_modules_0_modules_bn_parameters_bias_ = L_self_modules_features_modules_18_modules_branch0_modules_0_modules_bn_parameters_bias_
+        l_self_modules_features_modules_18_modules_branch0_modules_1_modules_conv_parameters_weight_ = L_self_modules_features_modules_18_modules_branch0_modules_1_modules_conv_parameters_weight_
+        l_self_modules_features_modules_18_modules_branch0_modules_1_modules_bn_buffers_running_mean_ = L_self_modules_features_modules_18_modules_branch0_modules_1_modules_bn_buffers_running_mean_
+        l_self_modules_features_modules_18_modules_branch0_modules_1_modules_bn_buffers_running_var_ = L_self_modules_features_modules_18_modules_branch0_modules_1_modules_bn_buffers_running_var_
+        l_self_modules_features_modules_18_modules_branch0_modules_1_modules_bn_parameters_weight_ = L_self_modules_features_modules_18_modules_branch0_modules_1_modules_bn_parameters_weight_
+        l_self_modules_features_modules_18_modules_branch0_modules_1_modules_bn_parameters_bias_ = L_self_modules_features_modules_18_modules_branch0_modules_1_modules_bn_parameters_bias_
+        l_self_modules_features_modules_18_modules_branch1_modules_0_modules_conv_parameters_weight_ = L_self_modules_features_modules_18_modules_branch1_modules_0_modules_conv_parameters_weight_
+        l_self_modules_features_modules_18_modules_branch1_modules_0_modules_bn_buffers_running_mean_ = L_self_modules_features_modules_18_modules_branch1_modules_0_modules_bn_buffers_running_mean_
+        l_self_modules_features_modules_18_modules_branch1_modules_0_modules_bn_buffers_running_var_ = L_self_modules_features_modules_18_modules_branch1_modules_0_modules_bn_buffers_running_var_
+        l_self_modules_features_modules_18_modules_branch1_modules_0_modules_bn_parameters_weight_ = L_self_modules_features_modules_18_modules_branch1_modules_0_modules_bn_parameters_weight_
+        l_self_modules_features_modules_18_modules_branch1_modules_0_modules_bn_parameters_bias_ = L_self_modules_features_modules_18_modules_branch1_modules_0_modules_bn_parameters_bias_
+        l_self_modules_features_modules_18_modules_branch1_modules_1_modules_conv_parameters_weight_ = L_self_modules_features_modules_18_modules_branch1_modules_1_modules_conv_parameters_weight_
+        l_self_modules_features_modules_18_modules_branch1_modules_1_modules_bn_buffers_running_mean_ = L_self_modules_features_modules_18_modules_branch1_modules_1_modules_bn_buffers_running_mean_
+        l_self_modules_features_modules_18_modules_branch1_modules_1_modules_bn_buffers_running_var_ = L_self_modules_features_modules_18_modules_branch1_modules_1_modules_bn_buffers_running_var_
+        l_self_modules_features_modules_18_modules_branch1_modules_1_modules_bn_parameters_weight_ = L_self_modules_features_modules_18_modules_branch1_modules_1_modules_bn_parameters_weight_
+        l_self_modules_features_modules_18_modules_branch1_modules_1_modules_bn_parameters_bias_ = L_self_modules_features_modules_18_modules_branch1_modules_1_modules_bn_parameters_bias_
+        l_self_modules_features_modules_18_modules_branch1_modules_2_modules_conv_parameters_weight_ = L_self_modules_features_modules_18_modules_branch1_modules_2_modules_conv_parameters_weight_
+        l_self_modules_features_modules_18_modules_branch1_modules_2_modules_bn_buffers_running_mean_ = L_self_modules_features_modules_18_modules_branch1_modules_2_modules_bn_buffers_running_mean_
+        l_self_modules_features_modules_18_modules_branch1_modules_2_modules_bn_buffers_running_var_ = L_self_modules_features_modules_18_modules_branch1_modules_2_modules_bn_buffers_running_var_
+        l_self_modules_features_modules_18_modules_branch1_modules_2_modules_bn_parameters_weight_ = L_self_modules_features_modules_18_modules_branch1_modules_2_modules_bn_parameters_weight_
+        l_self_modules_features_modules_18_modules_branch1_modules_2_modules_bn_parameters_bias_ = L_self_modules_features_modules_18_modules_branch1_modules_2_modules_bn_parameters_bias_
+        l_self_modules_features_modules_18_modules_branch1_modules_3_modules_conv_parameters_weight_ = L_self_modules_features_modules_18_modules_branch1_modules_3_modules_conv_parameters_weight_
+        l_self_modules_features_modules_18_modules_branch1_modules_3_modules_bn_buffers_running_mean_ = L_self_modules_features_modules_18_modules_branch1_modules_3_modules_bn_buffers_running_mean_
+        l_self_modules_features_modules_18_modules_branch1_modules_3_modules_bn_buffers_running_var_ = L_self_modules_features_modules_18_modules_branch1_modules_3_modules_bn_buffers_running_var_
+        l_self_modules_features_modules_18_modules_branch1_modules_3_modules_bn_parameters_weight_ = L_self_modules_features_modules_18_modules_branch1_modules_3_modules_bn_parameters_weight_
+        l_self_modules_features_modules_18_modules_branch1_modules_3_modules_bn_parameters_bias_ = L_self_modules_features_modules_18_modules_branch1_modules_3_modules_bn_parameters_bias_
+        l_self_modules_features_modules_19_modules_branch0_modules_conv_parameters_weight_ = L_self_modules_features_modules_19_modules_branch0_modules_conv_parameters_weight_
+        l_self_modules_features_modules_19_modules_branch0_modules_bn_buffers_running_mean_ = L_self_modules_features_modules_19_modules_branch0_modules_bn_buffers_running_mean_
+        l_self_modules_features_modules_19_modules_branch0_modules_bn_buffers_running_var_ = L_self_modules_features_modules_19_modules_branch0_modules_bn_buffers_running_var_
+        l_self_modules_features_modules_19_modules_branch0_modules_bn_parameters_weight_ = L_self_modules_features_modules_19_modules_branch0_modules_bn_parameters_weight_
+        l_self_modules_features_modules_19_modules_branch0_modules_bn_parameters_bias_ = L_self_modules_features_modules_19_modules_branch0_modules_bn_parameters_bias_
+        l_self_modules_features_modules_19_modules_branch1_0_modules_conv_parameters_weight_ = L_self_modules_features_modules_19_modules_branch1_0_modules_conv_parameters_weight_
+        l_self_modules_features_modules_19_modules_branch1_0_modules_bn_buffers_running_mean_ = L_self_modules_features_modules_19_modules_branch1_0_modules_bn_buffers_running_mean_
+        l_self_modules_features_modules_19_modules_branch1_0_modules_bn_buffers_running_var_ = L_self_modules_features_modules_19_modules_branch1_0_modules_bn_buffers_running_var_
+        l_self_modules_features_modules_19_modules_branch1_0_modules_bn_parameters_weight_ = L_self_modules_features_modules_19_modules_branch1_0_modules_bn_parameters_weight_
+        l_self_modules_features_modules_19_modules_branch1_0_modules_bn_parameters_bias_ = L_self_modules_features_modules_19_modules_branch1_0_modules_bn_parameters_bias_
+        l_self_modules_features_modules_19_modules_branch1_1a_modules_conv_parameters_weight_ = L_self_modules_features_modules_19_modules_branch1_1a_modules_conv_parameters_weight_
+        l_self_modules_features_modules_19_modules_branch1_1a_modules_bn_buffers_running_mean_ = L_self_modules_features_modules_19_modules_branch1_1a_modules_bn_buffers_running_mean_
+        l_self_modules_features_modules_19_modules_branch1_1a_modules_bn_buffers_running_var_ = L_self_modules_features_modules_19_modules_branch1_1a_modules_bn_buffers_running_var_
+        l_self_modules_features_modules_19_modules_branch1_1a_modules_bn_parameters_weight_ = L_self_modules_features_modules_19_modules_branch1_1a_modules_bn_parameters_weight_
+        l_self_modules_features_modules_19_modules_branch1_1a_modules_bn_parameters_bias_ = L_self_modules_features_modules_19_modules_branch1_1a_modules_bn_parameters_bias_
+        l_self_modules_features_modules_19_modules_branch1_1b_modules_conv_parameters_weight_ = L_self_modules_features_modules_19_modules_branch1_1b_modules_conv_parameters_weight_
+        l_self_modules_features_modules_19_modules_branch1_1b_modules_bn_buffers_running_mean_ = L_self_modules_features_modules_19_modules_branch1_1b_modules_bn_buffers_running_mean_
+        l_self_modules_features_modules_19_modules_branch1_1b_modules_bn_buffers_running_var_ = L_self_modules_features_modules_19_modules_branch1_1b_modules_bn_buffers_running_var_
+        l_self_modules_features_modules_19_modules_branch1_1b_modules_bn_parameters_weight_ = L_self_modules_features_modules_19_modules_branch1_1b_modules_bn_parameters_weight_
+        l_self_modules_features_modules_19_modules_branch1_1b_modules_bn_parameters_bias_ = L_self_modules_features_modules_19_modules_branch1_1b_modules_bn_parameters_bias_
+        l_self_modules_features_modules_19_modules_branch2_0_modules_conv_parameters_weight_ = L_self_modules_features_modules_19_modules_branch2_0_modules_conv_parameters_weight_
+        l_self_modules_features_modules_19_modules_branch2_0_modules_bn_buffers_running_mean_ = L_self_modules_features_modules_19_modules_branch2_0_modules_bn_buffers_running_mean_
+        l_self_modules_features_modules_19_modules_branch2_0_modules_bn_buffers_running_var_ = L_self_modules_features_modules_19_modules_branch2_0_modules_bn_buffers_running_var_
+        l_self_modules_features_modules_19_modules_branch2_0_modules_bn_parameters_weight_ = L_self_modules_features_modules_19_modules_branch2_0_modules_bn_parameters_weight_
+        l_self_modules_features_modules_19_modules_branch2_0_modules_bn_parameters_bias_ = L_self_modules_features_modules_19_modules_branch2_0_modules_bn_parameters_bias_
+        l_self_modules_features_modules_19_modules_branch2_1_modules_conv_parameters_weight_ = L_self_modules_features_modules_19_modules_branch2_1_modules_conv_parameters_weight_
+        l_self_modules_features_modules_19_modules_branch2_1_modules_bn_buffers_running_mean_ = L_self_modules_features_modules_19_modules_branch2_1_modules_bn_buffers_running_mean_
+        l_self_modules_features_modules_19_modules_branch2_1_modules_bn_buffers_running_var_ = L_self_modules_features_modules_19_modules_branch2_1_modules_bn_buffers_running_var_
+        l_self_modules_features_modules_19_modules_branch2_1_modules_bn_parameters_weight_ = L_self_modules_features_modules_19_modules_branch2_1_modules_bn_parameters_weight_
+        l_self_modules_features_modules_19_modules_branch2_1_modules_bn_parameters_bias_ = L_self_modules_features_modules_19_modules_branch2_1_modules_bn_parameters_bias_
+        l_self_modules_features_modules_19_modules_branch2_2_modules_conv_parameters_weight_ = L_self_modules_features_modules_19_modules_branch2_2_modules_conv_parameters_weight_
+        l_self_modules_features_modules_19_modules_branch2_2_modules_bn_buffers_running_mean_ = L_self_modules_features_modules_19_modules_branch2_2_modules_bn_buffers_running_mean_
+        l_self_modules_features_modules_19_modules_branch2_2_modules_bn_buffers_running_var_ = L_self_modules_features_modules_19_modules_branch2_2_modules_bn_buffers_running_var_
+        l_self_modules_features_modules_19_modules_branch2_2_modules_bn_parameters_weight_ = L_self_modules_features_modules_19_modules_branch2_2_modules_bn_parameters_weight_
+        l_self_modules_features_modules_19_modules_branch2_2_modules_bn_parameters_bias_ = L_self_modules_features_modules_19_modules_branch2_2_modules_bn_parameters_bias_
+        l_self_modules_features_modules_19_modules_branch2_3a_modules_conv_parameters_weight_ = L_self_modules_features_modules_19_modules_branch2_3a_modules_conv_parameters_weight_
+        l_self_modules_features_modules_19_modules_branch2_3a_modules_bn_buffers_running_mean_ = L_self_modules_features_modules_19_modules_branch2_3a_modules_bn_buffers_running_mean_
+        l_self_modules_features_modules_19_modules_branch2_3a_modules_bn_buffers_running_var_ = L_self_modules_features_modules_19_modules_branch2_3a_modules_bn_buffers_running_var_
+        l_self_modules_features_modules_19_modules_branch2_3a_modules_bn_parameters_weight_ = L_self_modules_features_modules_19_modules_branch2_3a_modules_bn_parameters_weight_
+        l_self_modules_features_modules_19_modules_branch2_3a_modules_bn_parameters_bias_ = L_self_modules_features_modules_19_modules_branch2_3a_modules_bn_parameters_bias_
+        l_self_modules_features_modules_19_modules_branch2_3b_modules_conv_parameters_weight_ = L_self_modules_features_modules_19_modules_branch2_3b_modules_conv_parameters_weight_
+        l_self_modules_features_modules_19_modules_branch2_3b_modules_bn_buffers_running_mean_ = L_self_modules_features_modules_19_modules_branch2_3b_modules_bn_buffers_running_mean_
+        l_self_modules_features_modules_19_modules_branch2_3b_modules_bn_buffers_running_var_ = L_self_modules_features_modules_19_modules_branch2_3b_modules_bn_buffers_running_var_
+        l_self_modules_features_modules_19_modules_branch2_3b_modules_bn_parameters_weight_ = L_self_modules_features_modules_19_modules_branch2_3b_modules_bn_parameters_weight_
+        l_self_modules_features_modules_19_modules_branch2_3b_modules_bn_parameters_bias_ = L_self_modules_features_modules_19_modules_branch2_3b_modules_bn_parameters_bias_
+        l_self_modules_features_modules_19_modules_branch3_modules_1_modules_conv_parameters_weight_ = L_self_modules_features_modules_19_modules_branch3_modules_1_modules_conv_parameters_weight_
+        l_self_modules_features_modules_19_modules_branch3_modules_1_modules_bn_buffers_running_mean_ = L_self_modules_features_modules_19_modules_branch3_modules_1_modules_bn_buffers_running_mean_
+        l_self_modules_features_modules_19_modules_branch3_modules_1_modules_bn_buffers_running_var_ = L_self_modules_features_modules_19_modules_branch3_modules_1_modules_bn_buffers_running_var_
+        l_self_modules_features_modules_19_modules_branch3_modules_1_modules_bn_parameters_weight_ = L_self_modules_features_modules_19_modules_branch3_modules_1_modules_bn_parameters_weight_
+        l_self_modules_features_modules_19_modules_branch3_modules_1_modules_bn_parameters_bias_ = L_self_modules_features_modules_19_modules_branch3_modules_1_modules_bn_parameters_bias_
+        l_self_modules_features_modules_20_modules_branch0_modules_conv_parameters_weight_ = L_self_modules_features_modules_20_modules_branch0_modules_conv_parameters_weight_
+        l_self_modules_features_modules_20_modules_branch0_modules_bn_buffers_running_mean_ = L_self_modules_features_modules_20_modules_branch0_modules_bn_buffers_running_mean_
+        l_self_modules_features_modules_20_modules_branch0_modules_bn_buffers_running_var_ = L_self_modules_features_modules_20_modules_branch0_modules_bn_buffers_running_var_
+        l_self_modules_features_modules_20_modules_branch0_modules_bn_parameters_weight_ = L_self_modules_features_modules_20_modules_branch0_modules_bn_parameters_weight_
+        l_self_modules_features_modules_20_modules_branch0_modules_bn_parameters_bias_ = L_self_modules_features_modules_20_modules_branch0_modules_bn_parameters_bias_
+        l_self_modules_features_modules_20_modules_branch1_0_modules_conv_parameters_weight_ = L_self_modules_features_modules_20_modules_branch1_0_modules_conv_parameters_weight_
+        l_self_modules_features_modules_20_modules_branch1_0_modules_bn_buffers_running_mean_ = L_self_modules_features_modules_20_modules_branch1_0_modules_bn_buffers_running_mean_
+        l_self_modules_features_modules_20_modules_branch1_0_modules_bn_buffers_running_var_ = L_self_modules_features_modules_20_modules_branch1_0_modules_bn_buffers_running_var_
+        l_self_modules_features_modules_20_modules_branch1_0_modules_bn_parameters_weight_ = L_self_modules_features_modules_20_modules_branch1_0_modules_bn_parameters_weight_
+        l_self_modules_features_modules_20_modules_branch1_0_modules_bn_parameters_bias_ = L_self_modules_features_modules_20_modules_branch1_0_modules_bn_parameters_bias_
+        l_self_modules_features_modules_20_modules_branch1_1a_modules_conv_parameters_weight_ = L_self_modules_features_modules_20_modules_branch1_1a_modules_conv_parameters_weight_
+        l_self_modules_features_modules_20_modules_branch1_1a_modules_bn_buffers_running_mean_ = L_self_modules_features_modules_20_modules_branch1_1a_modules_bn_buffers_running_mean_
+        l_self_modules_features_modules_20_modules_branch1_1a_modules_bn_buffers_running_var_ = L_self_modules_features_modules_20_modules_branch1_1a_modules_bn_buffers_running_var_
+        l_self_modules_features_modules_20_modules_branch1_1a_modules_bn_parameters_weight_ = L_self_modules_features_modules_20_modules_branch1_1a_modules_bn_parameters_weight_
+        l_self_modules_features_modules_20_modules_branch1_1a_modules_bn_parameters_bias_ = L_self_modules_features_modules_20_modules_branch1_1a_modules_bn_parameters_bias_
+        l_self_modules_features_modules_20_modules_branch1_1b_modules_conv_parameters_weight_ = L_self_modules_features_modules_20_modules_branch1_1b_modules_conv_parameters_weight_
+        l_self_modules_features_modules_20_modules_branch1_1b_modules_bn_buffers_running_mean_ = L_self_modules_features_modules_20_modules_branch1_1b_modules_bn_buffers_running_mean_
+        l_self_modules_features_modules_20_modules_branch1_1b_modules_bn_buffers_running_var_ = L_self_modules_features_modules_20_modules_branch1_1b_modules_bn_buffers_running_var_
+        l_self_modules_features_modules_20_modules_branch1_1b_modules_bn_parameters_weight_ = L_self_modules_features_modules_20_modules_branch1_1b_modules_bn_parameters_weight_
+        l_self_modules_features_modules_20_modules_branch1_1b_modules_bn_parameters_bias_ = L_self_modules_features_modules_20_modules_branch1_1b_modules_bn_parameters_bias_
+        l_self_modules_features_modules_20_modules_branch2_0_modules_conv_parameters_weight_ = L_self_modules_features_modules_20_modules_branch2_0_modules_conv_parameters_weight_
+        l_self_modules_features_modules_20_modules_branch2_0_modules_bn_buffers_running_mean_ = L_self_modules_features_modules_20_modules_branch2_0_modules_bn_buffers_running_mean_
+        l_self_modules_features_modules_20_modules_branch2_0_modules_bn_buffers_running_var_ = L_self_modules_features_modules_20_modules_branch2_0_modules_bn_buffers_running_var_
+        l_self_modules_features_modules_20_modules_branch2_0_modules_bn_parameters_weight_ = L_self_modules_features_modules_20_modules_branch2_0_modules_bn_parameters_weight_
+        l_self_modules_features_modules_20_modules_branch2_0_modules_bn_parameters_bias_ = L_self_modules_features_modules_20_modules_branch2_0_modules_bn_parameters_bias_
+        l_self_modules_features_modules_20_modules_branch2_1_modules_conv_parameters_weight_ = L_self_modules_features_modules_20_modules_branch2_1_modules_conv_parameters_weight_
+        l_self_modules_features_modules_20_modules_branch2_1_modules_bn_buffers_running_mean_ = L_self_modules_features_modules_20_modules_branch2_1_modules_bn_buffers_running_mean_
+        l_self_modules_features_modules_20_modules_branch2_1_modules_bn_buffers_running_var_ = L_self_modules_features_modules_20_modules_branch2_1_modules_bn_buffers_running_var_
+        l_self_modules_features_modules_20_modules_branch2_1_modules_bn_parameters_weight_ = L_self_modules_features_modules_20_modules_branch2_1_modules_bn_parameters_weight_
+        l_self_modules_features_modules_20_modules_branch2_1_modules_bn_parameters_bias_ = L_self_modules_features_modules_20_modules_branch2_1_modules_bn_parameters_bias_
+        l_self_modules_features_modules_20_modules_branch2_2_modules_conv_parameters_weight_ = L_self_modules_features_modules_20_modules_branch2_2_modules_conv_parameters_weight_
+        l_self_modules_features_modules_20_modules_branch2_2_modules_bn_buffers_running_mean_ = L_self_modules_features_modules_20_modules_branch2_2_modules_bn_buffers_running_mean_
+        l_self_modules_features_modules_20_modules_branch2_2_modules_bn_buffers_running_var_ = L_self_modules_features_modules_20_modules_branch2_2_modules_bn_buffers_running_var_
+        l_self_modules_features_modules_20_modules_branch2_2_modules_bn_parameters_weight_ = L_self_modules_features_modules_20_modules_branch2_2_modules_bn_parameters_weight_
+        l_self_modules_features_modules_20_modules_branch2_2_modules_bn_parameters_bias_ = L_self_modules_features_modules_20_modules_branch2_2_modules_bn_parameters_bias_
+        l_self_modules_features_modules_20_modules_branch2_3a_modules_conv_parameters_weight_ = L_self_modules_features_modules_20_modules_branch2_3a_modules_conv_parameters_weight_
+        l_self_modules_features_modules_20_modules_branch2_3a_modules_bn_buffers_running_mean_ = L_self_modules_features_modules_20_modules_branch2_3a_modules_bn_buffers_running_mean_
+        l_self_modules_features_modules_20_modules_branch2_3a_modules_bn_buffers_running_var_ = L_self_modules_features_modules_20_modules_branch2_3a_modules_bn_buffers_running_var_
+        l_self_modules_features_modules_20_modules_branch2_3a_modules_bn_parameters_weight_ = L_self_modules_features_modules_20_modules_branch2_3a_modules_bn_parameters_weight_
+        l_self_modules_features_modules_20_modules_branch2_3a_modules_bn_parameters_bias_ = L_self_modules_features_modules_20_modules_branch2_3a_modules_bn_parameters_bias_
+        l_self_modules_features_modules_20_modules_branch2_3b_modules_conv_parameters_weight_ = L_self_modules_features_modules_20_modules_branch2_3b_modules_conv_parameters_weight_
+        l_self_modules_features_modules_20_modules_branch2_3b_modules_bn_buffers_running_mean_ = L_self_modules_features_modules_20_modules_branch2_3b_modules_bn_buffers_running_mean_
+        l_self_modules_features_modules_20_modules_branch2_3b_modules_bn_buffers_running_var_ = L_self_modules_features_modules_20_modules_branch2_3b_modules_bn_buffers_running_var_
+        l_self_modules_features_modules_20_modules_branch2_3b_modules_bn_parameters_weight_ = L_self_modules_features_modules_20_modules_branch2_3b_modules_bn_parameters_weight_
+        l_self_modules_features_modules_20_modules_branch2_3b_modules_bn_parameters_bias_ = L_self_modules_features_modules_20_modules_branch2_3b_modules_bn_parameters_bias_
+        l_self_modules_features_modules_20_modules_branch3_modules_1_modules_conv_parameters_weight_ = L_self_modules_features_modules_20_modules_branch3_modules_1_modules_conv_parameters_weight_
+        l_self_modules_features_modules_20_modules_branch3_modules_1_modules_bn_buffers_running_mean_ = L_self_modules_features_modules_20_modules_branch3_modules_1_modules_bn_buffers_running_mean_
+        l_self_modules_features_modules_20_modules_branch3_modules_1_modules_bn_buffers_running_var_ = L_self_modules_features_modules_20_modules_branch3_modules_1_modules_bn_buffers_running_var_
+        l_self_modules_features_modules_20_modules_branch3_modules_1_modules_bn_parameters_weight_ = L_self_modules_features_modules_20_modules_branch3_modules_1_modules_bn_parameters_weight_
+        l_self_modules_features_modules_20_modules_branch3_modules_1_modules_bn_parameters_bias_ = L_self_modules_features_modules_20_modules_branch3_modules_1_modules_bn_parameters_bias_
+        l_self_modules_features_modules_21_modules_branch0_modules_conv_parameters_weight_ = L_self_modules_features_modules_21_modules_branch0_modules_conv_parameters_weight_
+        l_self_modules_features_modules_21_modules_branch0_modules_bn_buffers_running_mean_ = L_self_modules_features_modules_21_modules_branch0_modules_bn_buffers_running_mean_
+        l_self_modules_features_modules_21_modules_branch0_modules_bn_buffers_running_var_ = L_self_modules_features_modules_21_modules_branch0_modules_bn_buffers_running_var_
+        l_self_modules_features_modules_21_modules_branch0_modules_bn_parameters_weight_ = L_self_modules_features_modules_21_modules_branch0_modules_bn_parameters_weight_
+        l_self_modules_features_modules_21_modules_branch0_modules_bn_parameters_bias_ = L_self_modules_features_modules_21_modules_branch0_modules_bn_parameters_bias_
+        l_self_modules_features_modules_21_modules_branch1_0_modules_conv_parameters_weight_ = L_self_modules_features_modules_21_modules_branch1_0_modules_conv_parameters_weight_
+        l_self_modules_features_modules_21_modules_branch1_0_modules_bn_buffers_running_mean_ = L_self_modules_features_modules_21_modules_branch1_0_modules_bn_buffers_running_mean_
+        l_self_modules_features_modules_21_modules_branch1_0_modules_bn_buffers_running_var_ = L_self_modules_features_modules_21_modules_branch1_0_modules_bn_buffers_running_var_
+        l_self_modules_features_modules_21_modules_branch1_0_modules_bn_parameters_weight_ = L_self_modules_features_modules_21_modules_branch1_0_modules_bn_parameters_weight_
+        l_self_modules_features_modules_21_modules_branch1_0_modules_bn_parameters_bias_ = L_self_modules_features_modules_21_modules_branch1_0_modules_bn_parameters_bias_
+        l_self_modules_features_modules_21_modules_branch1_1a_modules_conv_parameters_weight_ = L_self_modules_features_modules_21_modules_branch1_1a_modules_conv_parameters_weight_
+        l_self_modules_features_modules_21_modules_branch1_1a_modules_bn_buffers_running_mean_ = L_self_modules_features_modules_21_modules_branch1_1a_modules_bn_buffers_running_mean_
+        l_self_modules_features_modules_21_modules_branch1_1a_modules_bn_buffers_running_var_ = L_self_modules_features_modules_21_modules_branch1_1a_modules_bn_buffers_running_var_
+        l_self_modules_features_modules_21_modules_branch1_1a_modules_bn_parameters_weight_ = L_self_modules_features_modules_21_modules_branch1_1a_modules_bn_parameters_weight_
+        l_self_modules_features_modules_21_modules_branch1_1a_modules_bn_parameters_bias_ = L_self_modules_features_modules_21_modules_branch1_1a_modules_bn_parameters_bias_
+        l_self_modules_features_modules_21_modules_branch1_1b_modules_conv_parameters_weight_ = L_self_modules_features_modules_21_modules_branch1_1b_modules_conv_parameters_weight_
+        l_self_modules_features_modules_21_modules_branch1_1b_modules_bn_buffers_running_mean_ = L_self_modules_features_modules_21_modules_branch1_1b_modules_bn_buffers_running_mean_
+        l_self_modules_features_modules_21_modules_branch1_1b_modules_bn_buffers_running_var_ = L_self_modules_features_modules_21_modules_branch1_1b_modules_bn_buffers_running_var_
+        l_self_modules_features_modules_21_modules_branch1_1b_modules_bn_parameters_weight_ = L_self_modules_features_modules_21_modules_branch1_1b_modules_bn_parameters_weight_
+        l_self_modules_features_modules_21_modules_branch1_1b_modules_bn_parameters_bias_ = L_self_modules_features_modules_21_modules_branch1_1b_modules_bn_parameters_bias_
+        l_self_modules_features_modules_21_modules_branch2_0_modules_conv_parameters_weight_ = L_self_modules_features_modules_21_modules_branch2_0_modules_conv_parameters_weight_
+        l_self_modules_features_modules_21_modules_branch2_0_modules_bn_buffers_running_mean_ = L_self_modules_features_modules_21_modules_branch2_0_modules_bn_buffers_running_mean_
+        l_self_modules_features_modules_21_modules_branch2_0_modules_bn_buffers_running_var_ = L_self_modules_features_modules_21_modules_branch2_0_modules_bn_buffers_running_var_
+        l_self_modules_features_modules_21_modules_branch2_0_modules_bn_parameters_weight_ = L_self_modules_features_modules_21_modules_branch2_0_modules_bn_parameters_weight_
+        l_self_modules_features_modules_21_modules_branch2_0_modules_bn_parameters_bias_ = L_self_modules_features_modules_21_modules_branch2_0_modules_bn_parameters_bias_
+        l_self_modules_features_modules_21_modules_branch2_1_modules_conv_parameters_weight_ = L_self_modules_features_modules_21_modules_branch2_1_modules_conv_parameters_weight_
+        l_self_modules_features_modules_21_modules_branch2_1_modules_bn_buffers_running_mean_ = L_self_modules_features_modules_21_modules_branch2_1_modules_bn_buffers_running_mean_
+        l_self_modules_features_modules_21_modules_branch2_1_modules_bn_buffers_running_var_ = L_self_modules_features_modules_21_modules_branch2_1_modules_bn_buffers_running_var_
+        l_self_modules_features_modules_21_modules_branch2_1_modules_bn_parameters_weight_ = L_self_modules_features_modules_21_modules_branch2_1_modules_bn_parameters_weight_
+        l_self_modules_features_modules_21_modules_branch2_1_modules_bn_parameters_bias_ = L_self_modules_features_modules_21_modules_branch2_1_modules_bn_parameters_bias_
+        l_self_modules_features_modules_21_modules_branch2_2_modules_conv_parameters_weight_ = L_self_modules_features_modules_21_modules_branch2_2_modules_conv_parameters_weight_
+        l_self_modules_features_modules_21_modules_branch2_2_modules_bn_buffers_running_mean_ = L_self_modules_features_modules_21_modules_branch2_2_modules_bn_buffers_running_mean_
+        l_self_modules_features_modules_21_modules_branch2_2_modules_bn_buffers_running_var_ = L_self_modules_features_modules_21_modules_branch2_2_modules_bn_buffers_running_var_
+        l_self_modules_features_modules_21_modules_branch2_2_modules_bn_parameters_weight_ = L_self_modules_features_modules_21_modules_branch2_2_modules_bn_parameters_weight_
+        l_self_modules_features_modules_21_modules_branch2_2_modules_bn_parameters_bias_ = L_self_modules_features_modules_21_modules_branch2_2_modules_bn_parameters_bias_
+        l_self_modules_features_modules_21_modules_branch2_3a_modules_conv_parameters_weight_ = L_self_modules_features_modules_21_modules_branch2_3a_modules_conv_parameters_weight_
+        l_self_modules_features_modules_21_modules_branch2_3a_modules_bn_buffers_running_mean_ = L_self_modules_features_modules_21_modules_branch2_3a_modules_bn_buffers_running_mean_
+        l_self_modules_features_modules_21_modules_branch2_3a_modules_bn_buffers_running_var_ = L_self_modules_features_modules_21_modules_branch2_3a_modules_bn_buffers_running_var_
+        l_self_modules_features_modules_21_modules_branch2_3a_modules_bn_parameters_weight_ = L_self_modules_features_modules_21_modules_branch2_3a_modules_bn_parameters_weight_
+        l_self_modules_features_modules_21_modules_branch2_3a_modules_bn_parameters_bias_ = L_self_modules_features_modules_21_modules_branch2_3a_modules_bn_parameters_bias_
+        l_self_modules_features_modules_21_modules_branch2_3b_modules_conv_parameters_weight_ = L_self_modules_features_modules_21_modules_branch2_3b_modules_conv_parameters_weight_
+        l_self_modules_features_modules_21_modules_branch2_3b_modules_bn_buffers_running_mean_ = L_self_modules_features_modules_21_modules_branch2_3b_modules_bn_buffers_running_mean_
+        l_self_modules_features_modules_21_modules_branch2_3b_modules_bn_buffers_running_var_ = L_self_modules_features_modules_21_modules_branch2_3b_modules_bn_buffers_running_var_
+        l_self_modules_features_modules_21_modules_branch2_3b_modules_bn_parameters_weight_ = L_self_modules_features_modules_21_modules_branch2_3b_modules_bn_parameters_weight_
+        l_self_modules_features_modules_21_modules_branch2_3b_modules_bn_parameters_bias_ = L_self_modules_features_modules_21_modules_branch2_3b_modules_bn_parameters_bias_
+        l_self_modules_features_modules_21_modules_branch3_modules_1_modules_conv_parameters_weight_ = L_self_modules_features_modules_21_modules_branch3_modules_1_modules_conv_parameters_weight_
+        l_self_modules_features_modules_21_modules_branch3_modules_1_modules_bn_buffers_running_mean_ = L_self_modules_features_modules_21_modules_branch3_modules_1_modules_bn_buffers_running_mean_
+        l_self_modules_features_modules_21_modules_branch3_modules_1_modules_bn_buffers_running_var_ = L_self_modules_features_modules_21_modules_branch3_modules_1_modules_bn_buffers_running_var_
+        l_self_modules_features_modules_21_modules_branch3_modules_1_modules_bn_parameters_weight_ = L_self_modules_features_modules_21_modules_branch3_modules_1_modules_bn_parameters_weight_
+        l_self_modules_features_modules_21_modules_branch3_modules_1_modules_bn_parameters_bias_ = L_self_modules_features_modules_21_modules_branch3_modules_1_modules_bn_parameters_bias_
+        l_self_modules_last_linear_parameters_weight_ = (
+            L_self_modules_last_linear_parameters_weight_
+        )
+        l_self_modules_last_linear_parameters_bias_ = (
+            L_self_modules_last_linear_parameters_bias_
+        )
+        x = torch.conv2d(
+            l_x_,
+            l_self_modules_features_modules_0_modules_conv_parameters_weight_,
+            None,
+            (2, 2),
+            (0, 0),
+            (1, 1),
+            1,
+        )
+        l_x_ = l_self_modules_features_modules_0_modules_conv_parameters_weight_ = None
+        x_1 = torch.nn.functional.batch_norm(
+            x,
+            l_self_modules_features_modules_0_modules_bn_buffers_running_mean_,
+            l_self_modules_features_modules_0_modules_bn_buffers_running_var_,
+            l_self_modules_features_modules_0_modules_bn_parameters_weight_,
+            l_self_modules_features_modules_0_modules_bn_parameters_bias_,
+            False,
+            0.1,
+            0.001,
+        )
+        x = (
+            l_self_modules_features_modules_0_modules_bn_buffers_running_mean_
+        ) = (
+            l_self_modules_features_modules_0_modules_bn_buffers_running_var_
+        ) = (
+            l_self_modules_features_modules_0_modules_bn_parameters_weight_
+        ) = l_self_modules_features_modules_0_modules_bn_parameters_bias_ = None
+        x_2 = torch.nn.functional.relu(x_1, inplace=True)
+        x_1 = None
+        x_3 = torch.conv2d(
+            x_2,
+            l_self_modules_features_modules_1_modules_conv_parameters_weight_,
+            None,
+            (1, 1),
+            (0, 0),
+            (1, 1),
+            1,
+        )
+        x_2 = l_self_modules_features_modules_1_modules_conv_parameters_weight_ = None
+        x_4 = torch.nn.functional.batch_norm(
+            x_3,
+            l_self_modules_features_modules_1_modules_bn_buffers_running_mean_,
+            l_self_modules_features_modules_1_modules_bn_buffers_running_var_,
+            l_self_modules_features_modules_1_modules_bn_parameters_weight_,
+            l_self_modules_features_modules_1_modules_bn_parameters_bias_,
+            False,
+            0.1,
+            0.001,
+        )
+        x_3 = (
+            l_self_modules_features_modules_1_modules_bn_buffers_running_mean_
+        ) = (
+            l_self_modules_features_modules_1_modules_bn_buffers_running_var_
+        ) = (
+            l_self_modules_features_modules_1_modules_bn_parameters_weight_
+        ) = l_self_modules_features_modules_1_modules_bn_parameters_bias_ = None
+        x_5 = torch.nn.functional.relu(x_4, inplace=True)
+        x_4 = None
+        x_6 = torch.conv2d(
+            x_5,
+            l_self_modules_features_modules_2_modules_conv_parameters_weight_,
+            None,
+            (1, 1),
+            (1, 1),
+            (1, 1),
+            1,
+        )
+        x_5 = l_self_modules_features_modules_2_modules_conv_parameters_weight_ = None
+        x_7 = torch.nn.functional.batch_norm(
+            x_6,
+            l_self_modules_features_modules_2_modules_bn_buffers_running_mean_,
+            l_self_modules_features_modules_2_modules_bn_buffers_running_var_,
+            l_self_modules_features_modules_2_modules_bn_parameters_weight_,
+            l_self_modules_features_modules_2_modules_bn_parameters_bias_,
+            False,
+            0.1,
+            0.001,
+        )
+        x_6 = (
+            l_self_modules_features_modules_2_modules_bn_buffers_running_mean_
+        ) = (
+            l_self_modules_features_modules_2_modules_bn_buffers_running_var_
+        ) = (
+            l_self_modules_features_modules_2_modules_bn_parameters_weight_
+        ) = l_self_modules_features_modules_2_modules_bn_parameters_bias_ = None
+        x_8 = torch.nn.functional.relu(x_7, inplace=True)
+        x_7 = None
+        x0 = torch.nn.functional.max_pool2d(
+            x_8, 3, 2, 0, 1, ceil_mode=False, return_indices=False
+        )
+        x_9 = torch.conv2d(
+            x_8,
+            l_self_modules_features_modules_3_modules_conv_modules_conv_parameters_weight_,
+            None,
+            (2, 2),
+            (0, 0),
+            (1, 1),
+            1,
+        )
+        x_8 = l_self_modules_features_modules_3_modules_conv_modules_conv_parameters_weight_ = (None)
+        x_10 = torch.nn.functional.batch_norm(
+            x_9,
+            l_self_modules_features_modules_3_modules_conv_modules_bn_buffers_running_mean_,
+            l_self_modules_features_modules_3_modules_conv_modules_bn_buffers_running_var_,
+            l_self_modules_features_modules_3_modules_conv_modules_bn_parameters_weight_,
+            l_self_modules_features_modules_3_modules_conv_modules_bn_parameters_bias_,
+            False,
+            0.1,
+            0.001,
+        )
+        x_9 = l_self_modules_features_modules_3_modules_conv_modules_bn_buffers_running_mean_ = l_self_modules_features_modules_3_modules_conv_modules_bn_buffers_running_var_ = (
+            l_self_modules_features_modules_3_modules_conv_modules_bn_parameters_weight_
+        ) = (
+            l_self_modules_features_modules_3_modules_conv_modules_bn_parameters_bias_
+        ) = None
+        x_11 = torch.nn.functional.relu(x_10, inplace=True)
+        x_10 = None
+        out = torch.cat((x0, x_11), 1)
+        x0 = x_11 = None
+        x_12 = torch.conv2d(
+            out,
+            l_self_modules_features_modules_4_modules_branch0_modules_0_modules_conv_parameters_weight_,
+            None,
+            (1, 1),
+            (0, 0),
+            (1, 1),
+            1,
+        )
+        l_self_modules_features_modules_4_modules_branch0_modules_0_modules_conv_parameters_weight_ = (
+            None
+        )
+        x_13 = torch.nn.functional.batch_norm(
+            x_12,
+            l_self_modules_features_modules_4_modules_branch0_modules_0_modules_bn_buffers_running_mean_,
+            l_self_modules_features_modules_4_modules_branch0_modules_0_modules_bn_buffers_running_var_,
+            l_self_modules_features_modules_4_modules_branch0_modules_0_modules_bn_parameters_weight_,
+            l_self_modules_features_modules_4_modules_branch0_modules_0_modules_bn_parameters_bias_,
+            False,
+            0.1,
+            0.001,
+        )
+        x_12 = l_self_modules_features_modules_4_modules_branch0_modules_0_modules_bn_buffers_running_mean_ = l_self_modules_features_modules_4_modules_branch0_modules_0_modules_bn_buffers_running_var_ = l_self_modules_features_modules_4_modules_branch0_modules_0_modules_bn_parameters_weight_ = l_self_modules_features_modules_4_modules_branch0_modules_0_modules_bn_parameters_bias_ = (None)
+        x_14 = torch.nn.functional.relu(x_13, inplace=True)
+        x_13 = None
+        x_15 = torch.conv2d(
+            x_14,
+            l_self_modules_features_modules_4_modules_branch0_modules_1_modules_conv_parameters_weight_,
+            None,
+            (1, 1),
+            (0, 0),
+            (1, 1),
+            1,
+        )
+        x_14 = l_self_modules_features_modules_4_modules_branch0_modules_1_modules_conv_parameters_weight_ = (None)
+        x_16 = torch.nn.functional.batch_norm(
+            x_15,
+            l_self_modules_features_modules_4_modules_branch0_modules_1_modules_bn_buffers_running_mean_,
+            l_self_modules_features_modules_4_modules_branch0_modules_1_modules_bn_buffers_running_var_,
+            l_self_modules_features_modules_4_modules_branch0_modules_1_modules_bn_parameters_weight_,
+            l_self_modules_features_modules_4_modules_branch0_modules_1_modules_bn_parameters_bias_,
+            False,
+            0.1,
+            0.001,
+        )
+        x_15 = l_self_modules_features_modules_4_modules_branch0_modules_1_modules_bn_buffers_running_mean_ = l_self_modules_features_modules_4_modules_branch0_modules_1_modules_bn_buffers_running_var_ = l_self_modules_features_modules_4_modules_branch0_modules_1_modules_bn_parameters_weight_ = l_self_modules_features_modules_4_modules_branch0_modules_1_modules_bn_parameters_bias_ = (None)
+        x_17 = torch.nn.functional.relu(x_16, inplace=True)
+        x_16 = None
+        x_18 = torch.conv2d(
+            out,
+            l_self_modules_features_modules_4_modules_branch1_modules_0_modules_conv_parameters_weight_,
+            None,
+            (1, 1),
+            (0, 0),
+            (1, 1),
+            1,
+        )
+        out = l_self_modules_features_modules_4_modules_branch1_modules_0_modules_conv_parameters_weight_ = (None)
+        x_19 = torch.nn.functional.batch_norm(
+            x_18,
+            l_self_modules_features_modules_4_modules_branch1_modules_0_modules_bn_buffers_running_mean_,
+            l_self_modules_features_modules_4_modules_branch1_modules_0_modules_bn_buffers_running_var_,
+            l_self_modules_features_modules_4_modules_branch1_modules_0_modules_bn_parameters_weight_,
+            l_self_modules_features_modules_4_modules_branch1_modules_0_modules_bn_parameters_bias_,
+            False,
+            0.1,
+            0.001,
+        )
+        x_18 = l_self_modules_features_modules_4_modules_branch1_modules_0_modules_bn_buffers_running_mean_ = l_self_modules_features_modules_4_modules_branch1_modules_0_modules_bn_buffers_running_var_ = l_self_modules_features_modules_4_modules_branch1_modules_0_modules_bn_parameters_weight_ = l_self_modules_features_modules_4_modules_branch1_modules_0_modules_bn_parameters_bias_ = (None)
+        x_20 = torch.nn.functional.relu(x_19, inplace=True)
+        x_19 = None
+        x_21 = torch.conv2d(
+            x_20,
+            l_self_modules_features_modules_4_modules_branch1_modules_1_modules_conv_parameters_weight_,
+            None,
+            (1, 1),
+            (0, 3),
+            (1, 1),
+            1,
+        )
+        x_20 = l_self_modules_features_modules_4_modules_branch1_modules_1_modules_conv_parameters_weight_ = (None)
+        x_22 = torch.nn.functional.batch_norm(
+            x_21,
+            l_self_modules_features_modules_4_modules_branch1_modules_1_modules_bn_buffers_running_mean_,
+            l_self_modules_features_modules_4_modules_branch1_modules_1_modules_bn_buffers_running_var_,
+            l_self_modules_features_modules_4_modules_branch1_modules_1_modules_bn_parameters_weight_,
+            l_self_modules_features_modules_4_modules_branch1_modules_1_modules_bn_parameters_bias_,
+            False,
+            0.1,
+            0.001,
+        )
+        x_21 = l_self_modules_features_modules_4_modules_branch1_modules_1_modules_bn_buffers_running_mean_ = l_self_modules_features_modules_4_modules_branch1_modules_1_modules_bn_buffers_running_var_ = l_self_modules_features_modules_4_modules_branch1_modules_1_modules_bn_parameters_weight_ = l_self_modules_features_modules_4_modules_branch1_modules_1_modules_bn_parameters_bias_ = (None)
+        x_23 = torch.nn.functional.relu(x_22, inplace=True)
+        x_22 = None
+        x_24 = torch.conv2d(
+            x_23,
+            l_self_modules_features_modules_4_modules_branch1_modules_2_modules_conv_parameters_weight_,
+            None,
+            (1, 1),
+            (3, 0),
+            (1, 1),
+            1,
+        )
+        x_23 = l_self_modules_features_modules_4_modules_branch1_modules_2_modules_conv_parameters_weight_ = (None)
+        x_25 = torch.nn.functional.batch_norm(
+            x_24,
+            l_self_modules_features_modules_4_modules_branch1_modules_2_modules_bn_buffers_running_mean_,
+            l_self_modules_features_modules_4_modules_branch1_modules_2_modules_bn_buffers_running_var_,
+            l_self_modules_features_modules_4_modules_branch1_modules_2_modules_bn_parameters_weight_,
+            l_self_modules_features_modules_4_modules_branch1_modules_2_modules_bn_parameters_bias_,
+            False,
+            0.1,
+            0.001,
+        )
+        x_24 = l_self_modules_features_modules_4_modules_branch1_modules_2_modules_bn_buffers_running_mean_ = l_self_modules_features_modules_4_modules_branch1_modules_2_modules_bn_buffers_running_var_ = l_self_modules_features_modules_4_modules_branch1_modules_2_modules_bn_parameters_weight_ = l_self_modules_features_modules_4_modules_branch1_modules_2_modules_bn_parameters_bias_ = (None)
+        x_26 = torch.nn.functional.relu(x_25, inplace=True)
+        x_25 = None
+        x_27 = torch.conv2d(
+            x_26,
+            l_self_modules_features_modules_4_modules_branch1_modules_3_modules_conv_parameters_weight_,
+            None,
+            (1, 1),
+            (0, 0),
+            (1, 1),
+            1,
+        )
+        x_26 = l_self_modules_features_modules_4_modules_branch1_modules_3_modules_conv_parameters_weight_ = (None)
+        x_28 = torch.nn.functional.batch_norm(
+            x_27,
+            l_self_modules_features_modules_4_modules_branch1_modules_3_modules_bn_buffers_running_mean_,
+            l_self_modules_features_modules_4_modules_branch1_modules_3_modules_bn_buffers_running_var_,
+            l_self_modules_features_modules_4_modules_branch1_modules_3_modules_bn_parameters_weight_,
+            l_self_modules_features_modules_4_modules_branch1_modules_3_modules_bn_parameters_bias_,
+            False,
+            0.1,
+            0.001,
+        )
+        x_27 = l_self_modules_features_modules_4_modules_branch1_modules_3_modules_bn_buffers_running_mean_ = l_self_modules_features_modules_4_modules_branch1_modules_3_modules_bn_buffers_running_var_ = l_self_modules_features_modules_4_modules_branch1_modules_3_modules_bn_parameters_weight_ = l_self_modules_features_modules_4_modules_branch1_modules_3_modules_bn_parameters_bias_ = (None)
+        x_29 = torch.nn.functional.relu(x_28, inplace=True)
+        x_28 = None
+        out_1 = torch.cat((x_17, x_29), 1)
+        x_17 = x_29 = None
+        x_30 = torch.conv2d(
+            out_1,
+            l_self_modules_features_modules_5_modules_conv_modules_conv_parameters_weight_,
+            None,
+            (2, 2),
+            (0, 0),
+            (1, 1),
+            1,
+        )
+        l_self_modules_features_modules_5_modules_conv_modules_conv_parameters_weight_ = (
+            None
+        )
+        x_31 = torch.nn.functional.batch_norm(
+            x_30,
+            l_self_modules_features_modules_5_modules_conv_modules_bn_buffers_running_mean_,
+            l_self_modules_features_modules_5_modules_conv_modules_bn_buffers_running_var_,
+            l_self_modules_features_modules_5_modules_conv_modules_bn_parameters_weight_,
+            l_self_modules_features_modules_5_modules_conv_modules_bn_parameters_bias_,
+            False,
+            0.1,
+            0.001,
+        )
+        x_30 = l_self_modules_features_modules_5_modules_conv_modules_bn_buffers_running_mean_ = l_self_modules_features_modules_5_modules_conv_modules_bn_buffers_running_var_ = (
+            l_self_modules_features_modules_5_modules_conv_modules_bn_parameters_weight_
+        ) = (
+            l_self_modules_features_modules_5_modules_conv_modules_bn_parameters_bias_
+        ) = None
+        x_32 = torch.nn.functional.relu(x_31, inplace=True)
+        x_31 = None
+        x1 = torch.nn.functional.max_pool2d(
+            out_1, 3, 2, 0, 1, ceil_mode=False, return_indices=False
+        )
+        out_1 = None
+        out_2 = torch.cat((x_32, x1), 1)
+        x_32 = x1 = None
+        x_33 = torch.conv2d(
+            out_2,
+            l_self_modules_features_modules_6_modules_branch0_modules_conv_parameters_weight_,
+            None,
+            (1, 1),
+            (0, 0),
+            (1, 1),
+            1,
+        )
+        l_self_modules_features_modules_6_modules_branch0_modules_conv_parameters_weight_ = (
+            None
+        )
+        x_34 = torch.nn.functional.batch_norm(
+            x_33,
+            l_self_modules_features_modules_6_modules_branch0_modules_bn_buffers_running_mean_,
+            l_self_modules_features_modules_6_modules_branch0_modules_bn_buffers_running_var_,
+            l_self_modules_features_modules_6_modules_branch0_modules_bn_parameters_weight_,
+            l_self_modules_features_modules_6_modules_branch0_modules_bn_parameters_bias_,
+            False,
+            0.1,
+            0.001,
+        )
+        x_33 = l_self_modules_features_modules_6_modules_branch0_modules_bn_buffers_running_mean_ = l_self_modules_features_modules_6_modules_branch0_modules_bn_buffers_running_var_ = l_self_modules_features_modules_6_modules_branch0_modules_bn_parameters_weight_ = l_self_modules_features_modules_6_modules_branch0_modules_bn_parameters_bias_ = (None)
+        x_35 = torch.nn.functional.relu(x_34, inplace=True)
+        x_34 = None
+        x_36 = torch.conv2d(
+            out_2,
+            l_self_modules_features_modules_6_modules_branch1_modules_0_modules_conv_parameters_weight_,
+            None,
+            (1, 1),
+            (0, 0),
+            (1, 1),
+            1,
+        )
+        l_self_modules_features_modules_6_modules_branch1_modules_0_modules_conv_parameters_weight_ = (
+            None
+        )
+        x_37 = torch.nn.functional.batch_norm(
+            x_36,
+            l_self_modules_features_modules_6_modules_branch1_modules_0_modules_bn_buffers_running_mean_,
+            l_self_modules_features_modules_6_modules_branch1_modules_0_modules_bn_buffers_running_var_,
+            l_self_modules_features_modules_6_modules_branch1_modules_0_modules_bn_parameters_weight_,
+            l_self_modules_features_modules_6_modules_branch1_modules_0_modules_bn_parameters_bias_,
+            False,
+            0.1,
+            0.001,
+        )
+        x_36 = l_self_modules_features_modules_6_modules_branch1_modules_0_modules_bn_buffers_running_mean_ = l_self_modules_features_modules_6_modules_branch1_modules_0_modules_bn_buffers_running_var_ = l_self_modules_features_modules_6_modules_branch1_modules_0_modules_bn_parameters_weight_ = l_self_modules_features_modules_6_modules_branch1_modules_0_modules_bn_parameters_bias_ = (None)
+        x_38 = torch.nn.functional.relu(x_37, inplace=True)
+        x_37 = None
+        x_39 = torch.conv2d(
+            x_38,
+            l_self_modules_features_modules_6_modules_branch1_modules_1_modules_conv_parameters_weight_,
+            None,
+            (1, 1),
+            (1, 1),
+            (1, 1),
+            1,
+        )
+        x_38 = l_self_modules_features_modules_6_modules_branch1_modules_1_modules_conv_parameters_weight_ = (None)
+        x_40 = torch.nn.functional.batch_norm(
+            x_39,
+            l_self_modules_features_modules_6_modules_branch1_modules_1_modules_bn_buffers_running_mean_,
+            l_self_modules_features_modules_6_modules_branch1_modules_1_modules_bn_buffers_running_var_,
+            l_self_modules_features_modules_6_modules_branch1_modules_1_modules_bn_parameters_weight_,
+            l_self_modules_features_modules_6_modules_branch1_modules_1_modules_bn_parameters_bias_,
+            False,
+            0.1,
+            0.001,
+        )
+        x_39 = l_self_modules_features_modules_6_modules_branch1_modules_1_modules_bn_buffers_running_mean_ = l_self_modules_features_modules_6_modules_branch1_modules_1_modules_bn_buffers_running_var_ = l_self_modules_features_modules_6_modules_branch1_modules_1_modules_bn_parameters_weight_ = l_self_modules_features_modules_6_modules_branch1_modules_1_modules_bn_parameters_bias_ = (None)
+        x_41 = torch.nn.functional.relu(x_40, inplace=True)
+        x_40 = None
+        x_42 = torch.conv2d(
+            out_2,
+            l_self_modules_features_modules_6_modules_branch2_modules_0_modules_conv_parameters_weight_,
+            None,
+            (1, 1),
+            (0, 0),
+            (1, 1),
+            1,
+        )
+        l_self_modules_features_modules_6_modules_branch2_modules_0_modules_conv_parameters_weight_ = (
+            None
+        )
+        x_43 = torch.nn.functional.batch_norm(
+            x_42,
+            l_self_modules_features_modules_6_modules_branch2_modules_0_modules_bn_buffers_running_mean_,
+            l_self_modules_features_modules_6_modules_branch2_modules_0_modules_bn_buffers_running_var_,
+            l_self_modules_features_modules_6_modules_branch2_modules_0_modules_bn_parameters_weight_,
+            l_self_modules_features_modules_6_modules_branch2_modules_0_modules_bn_parameters_bias_,
+            False,
+            0.1,
+            0.001,
+        )
+        x_42 = l_self_modules_features_modules_6_modules_branch2_modules_0_modules_bn_buffers_running_mean_ = l_self_modules_features_modules_6_modules_branch2_modules_0_modules_bn_buffers_running_var_ = l_self_modules_features_modules_6_modules_branch2_modules_0_modules_bn_parameters_weight_ = l_self_modules_features_modules_6_modules_branch2_modules_0_modules_bn_parameters_bias_ = (None)
+        x_44 = torch.nn.functional.relu(x_43, inplace=True)
+        x_43 = None
+        x_45 = torch.conv2d(
+            x_44,
+            l_self_modules_features_modules_6_modules_branch2_modules_1_modules_conv_parameters_weight_,
+            None,
+            (1, 1),
+            (1, 1),
+            (1, 1),
+            1,
+        )
+        x_44 = l_self_modules_features_modules_6_modules_branch2_modules_1_modules_conv_parameters_weight_ = (None)
+        x_46 = torch.nn.functional.batch_norm(
+            x_45,
+            l_self_modules_features_modules_6_modules_branch2_modules_1_modules_bn_buffers_running_mean_,
+            l_self_modules_features_modules_6_modules_branch2_modules_1_modules_bn_buffers_running_var_,
+            l_self_modules_features_modules_6_modules_branch2_modules_1_modules_bn_parameters_weight_,
+            l_self_modules_features_modules_6_modules_branch2_modules_1_modules_bn_parameters_bias_,
+            False,
+            0.1,
+            0.001,
+        )
+        x_45 = l_self_modules_features_modules_6_modules_branch2_modules_1_modules_bn_buffers_running_mean_ = l_self_modules_features_modules_6_modules_branch2_modules_1_modules_bn_buffers_running_var_ = l_self_modules_features_modules_6_modules_branch2_modules_1_modules_bn_parameters_weight_ = l_self_modules_features_modules_6_modules_branch2_modules_1_modules_bn_parameters_bias_ = (None)
+        x_47 = torch.nn.functional.relu(x_46, inplace=True)
+        x_46 = None
+        x_48 = torch.conv2d(
+            x_47,
+            l_self_modules_features_modules_6_modules_branch2_modules_2_modules_conv_parameters_weight_,
+            None,
+            (1, 1),
+            (1, 1),
+            (1, 1),
+            1,
+        )
+        x_47 = l_self_modules_features_modules_6_modules_branch2_modules_2_modules_conv_parameters_weight_ = (None)
+        x_49 = torch.nn.functional.batch_norm(
+            x_48,
+            l_self_modules_features_modules_6_modules_branch2_modules_2_modules_bn_buffers_running_mean_,
+            l_self_modules_features_modules_6_modules_branch2_modules_2_modules_bn_buffers_running_var_,
+            l_self_modules_features_modules_6_modules_branch2_modules_2_modules_bn_parameters_weight_,
+            l_self_modules_features_modules_6_modules_branch2_modules_2_modules_bn_parameters_bias_,
+            False,
+            0.1,
+            0.001,
+        )
+        x_48 = l_self_modules_features_modules_6_modules_branch2_modules_2_modules_bn_buffers_running_mean_ = l_self_modules_features_modules_6_modules_branch2_modules_2_modules_bn_buffers_running_var_ = l_self_modules_features_modules_6_modules_branch2_modules_2_modules_bn_parameters_weight_ = l_self_modules_features_modules_6_modules_branch2_modules_2_modules_bn_parameters_bias_ = (None)
+        x_50 = torch.nn.functional.relu(x_49, inplace=True)
+        x_49 = None
+        input_1 = torch._C._nn.avg_pool2d(out_2, 3, 1, 1, False, False, None)
+        out_2 = None
+        x_51 = torch.conv2d(
+            input_1,
+            l_self_modules_features_modules_6_modules_branch3_modules_1_modules_conv_parameters_weight_,
+            None,
+            (1, 1),
+            (0, 0),
+            (1, 1),
+            1,
+        )
+        input_1 = l_self_modules_features_modules_6_modules_branch3_modules_1_modules_conv_parameters_weight_ = (None)
+        x_52 = torch.nn.functional.batch_norm(
+            x_51,
+            l_self_modules_features_modules_6_modules_branch3_modules_1_modules_bn_buffers_running_mean_,
+            l_self_modules_features_modules_6_modules_branch3_modules_1_modules_bn_buffers_running_var_,
+            l_self_modules_features_modules_6_modules_branch3_modules_1_modules_bn_parameters_weight_,
+            l_self_modules_features_modules_6_modules_branch3_modules_1_modules_bn_parameters_bias_,
+            False,
+            0.1,
+            0.001,
+        )
+        x_51 = l_self_modules_features_modules_6_modules_branch3_modules_1_modules_bn_buffers_running_mean_ = l_self_modules_features_modules_6_modules_branch3_modules_1_modules_bn_buffers_running_var_ = l_self_modules_features_modules_6_modules_branch3_modules_1_modules_bn_parameters_weight_ = l_self_modules_features_modules_6_modules_branch3_modules_1_modules_bn_parameters_bias_ = (None)
+        x_53 = torch.nn.functional.relu(x_52, inplace=True)
+        x_52 = None
+        out_3 = torch.cat((x_35, x_41, x_50, x_53), 1)
+        x_35 = x_41 = x_50 = x_53 = None
+        x_54 = torch.conv2d(
+            out_3,
+            l_self_modules_features_modules_7_modules_branch0_modules_conv_parameters_weight_,
+            None,
+            (1, 1),
+            (0, 0),
+            (1, 1),
+            1,
+        )
+        l_self_modules_features_modules_7_modules_branch0_modules_conv_parameters_weight_ = (
+            None
+        )
+        x_55 = torch.nn.functional.batch_norm(
+            x_54,
+            l_self_modules_features_modules_7_modules_branch0_modules_bn_buffers_running_mean_,
+            l_self_modules_features_modules_7_modules_branch0_modules_bn_buffers_running_var_,
+            l_self_modules_features_modules_7_modules_branch0_modules_bn_parameters_weight_,
+            l_self_modules_features_modules_7_modules_branch0_modules_bn_parameters_bias_,
+            False,
+            0.1,
+            0.001,
+        )
+        x_54 = l_self_modules_features_modules_7_modules_branch0_modules_bn_buffers_running_mean_ = l_self_modules_features_modules_7_modules_branch0_modules_bn_buffers_running_var_ = l_self_modules_features_modules_7_modules_branch0_modules_bn_parameters_weight_ = l_self_modules_features_modules_7_modules_branch0_modules_bn_parameters_bias_ = (None)
+        x_56 = torch.nn.functional.relu(x_55, inplace=True)
+        x_55 = None
+        x_57 = torch.conv2d(
+            out_3,
+            l_self_modules_features_modules_7_modules_branch1_modules_0_modules_conv_parameters_weight_,
+            None,
+            (1, 1),
+            (0, 0),
+            (1, 1),
+            1,
+        )
+        l_self_modules_features_modules_7_modules_branch1_modules_0_modules_conv_parameters_weight_ = (
+            None
+        )
+        x_58 = torch.nn.functional.batch_norm(
+            x_57,
+            l_self_modules_features_modules_7_modules_branch1_modules_0_modules_bn_buffers_running_mean_,
+            l_self_modules_features_modules_7_modules_branch1_modules_0_modules_bn_buffers_running_var_,
+            l_self_modules_features_modules_7_modules_branch1_modules_0_modules_bn_parameters_weight_,
+            l_self_modules_features_modules_7_modules_branch1_modules_0_modules_bn_parameters_bias_,
+            False,
+            0.1,
+            0.001,
+        )
+        x_57 = l_self_modules_features_modules_7_modules_branch1_modules_0_modules_bn_buffers_running_mean_ = l_self_modules_features_modules_7_modules_branch1_modules_0_modules_bn_buffers_running_var_ = l_self_modules_features_modules_7_modules_branch1_modules_0_modules_bn_parameters_weight_ = l_self_modules_features_modules_7_modules_branch1_modules_0_modules_bn_parameters_bias_ = (None)
+        x_59 = torch.nn.functional.relu(x_58, inplace=True)
+        x_58 = None
+        x_60 = torch.conv2d(
+            x_59,
+            l_self_modules_features_modules_7_modules_branch1_modules_1_modules_conv_parameters_weight_,
+            None,
+            (1, 1),
+            (1, 1),
+            (1, 1),
+            1,
+        )
+        x_59 = l_self_modules_features_modules_7_modules_branch1_modules_1_modules_conv_parameters_weight_ = (None)
+        x_61 = torch.nn.functional.batch_norm(
+            x_60,
+            l_self_modules_features_modules_7_modules_branch1_modules_1_modules_bn_buffers_running_mean_,
+            l_self_modules_features_modules_7_modules_branch1_modules_1_modules_bn_buffers_running_var_,
+            l_self_modules_features_modules_7_modules_branch1_modules_1_modules_bn_parameters_weight_,
+            l_self_modules_features_modules_7_modules_branch1_modules_1_modules_bn_parameters_bias_,
+            False,
+            0.1,
+            0.001,
+        )
+        x_60 = l_self_modules_features_modules_7_modules_branch1_modules_1_modules_bn_buffers_running_mean_ = l_self_modules_features_modules_7_modules_branch1_modules_1_modules_bn_buffers_running_var_ = l_self_modules_features_modules_7_modules_branch1_modules_1_modules_bn_parameters_weight_ = l_self_modules_features_modules_7_modules_branch1_modules_1_modules_bn_parameters_bias_ = (None)
+        x_62 = torch.nn.functional.relu(x_61, inplace=True)
+        x_61 = None
+        x_63 = torch.conv2d(
+            out_3,
+            l_self_modules_features_modules_7_modules_branch2_modules_0_modules_conv_parameters_weight_,
+            None,
+            (1, 1),
+            (0, 0),
+            (1, 1),
+            1,
+        )
+        l_self_modules_features_modules_7_modules_branch2_modules_0_modules_conv_parameters_weight_ = (
+            None
+        )
+        x_64 = torch.nn.functional.batch_norm(
+            x_63,
+            l_self_modules_features_modules_7_modules_branch2_modules_0_modules_bn_buffers_running_mean_,
+            l_self_modules_features_modules_7_modules_branch2_modules_0_modules_bn_buffers_running_var_,
+            l_self_modules_features_modules_7_modules_branch2_modules_0_modules_bn_parameters_weight_,
+            l_self_modules_features_modules_7_modules_branch2_modules_0_modules_bn_parameters_bias_,
+            False,
+            0.1,
+            0.001,
+        )
+        x_63 = l_self_modules_features_modules_7_modules_branch2_modules_0_modules_bn_buffers_running_mean_ = l_self_modules_features_modules_7_modules_branch2_modules_0_modules_bn_buffers_running_var_ = l_self_modules_features_modules_7_modules_branch2_modules_0_modules_bn_parameters_weight_ = l_self_modules_features_modules_7_modules_branch2_modules_0_modules_bn_parameters_bias_ = (None)
+        x_65 = torch.nn.functional.relu(x_64, inplace=True)
+        x_64 = None
+        x_66 = torch.conv2d(
+            x_65,
+            l_self_modules_features_modules_7_modules_branch2_modules_1_modules_conv_parameters_weight_,
+            None,
+            (1, 1),
+            (1, 1),
+            (1, 1),
+            1,
+        )
+        x_65 = l_self_modules_features_modules_7_modules_branch2_modules_1_modules_conv_parameters_weight_ = (None)
+        x_67 = torch.nn.functional.batch_norm(
+            x_66,
+            l_self_modules_features_modules_7_modules_branch2_modules_1_modules_bn_buffers_running_mean_,
+            l_self_modules_features_modules_7_modules_branch2_modules_1_modules_bn_buffers_running_var_,
+            l_self_modules_features_modules_7_modules_branch2_modules_1_modules_bn_parameters_weight_,
+            l_self_modules_features_modules_7_modules_branch2_modules_1_modules_bn_parameters_bias_,
+            False,
+            0.1,
+            0.001,
+        )
+        x_66 = l_self_modules_features_modules_7_modules_branch2_modules_1_modules_bn_buffers_running_mean_ = l_self_modules_features_modules_7_modules_branch2_modules_1_modules_bn_buffers_running_var_ = l_self_modules_features_modules_7_modules_branch2_modules_1_modules_bn_parameters_weight_ = l_self_modules_features_modules_7_modules_branch2_modules_1_modules_bn_parameters_bias_ = (None)
+        x_68 = torch.nn.functional.relu(x_67, inplace=True)
+        x_67 = None
+        x_69 = torch.conv2d(
+            x_68,
+            l_self_modules_features_modules_7_modules_branch2_modules_2_modules_conv_parameters_weight_,
+            None,
+            (1, 1),
+            (1, 1),
+            (1, 1),
+            1,
+        )
+        x_68 = l_self_modules_features_modules_7_modules_branch2_modules_2_modules_conv_parameters_weight_ = (None)
+        x_70 = torch.nn.functional.batch_norm(
+            x_69,
+            l_self_modules_features_modules_7_modules_branch2_modules_2_modules_bn_buffers_running_mean_,
+            l_self_modules_features_modules_7_modules_branch2_modules_2_modules_bn_buffers_running_var_,
+            l_self_modules_features_modules_7_modules_branch2_modules_2_modules_bn_parameters_weight_,
+            l_self_modules_features_modules_7_modules_branch2_modules_2_modules_bn_parameters_bias_,
+            False,
+            0.1,
+            0.001,
+        )
+        x_69 = l_self_modules_features_modules_7_modules_branch2_modules_2_modules_bn_buffers_running_mean_ = l_self_modules_features_modules_7_modules_branch2_modules_2_modules_bn_buffers_running_var_ = l_self_modules_features_modules_7_modules_branch2_modules_2_modules_bn_parameters_weight_ = l_self_modules_features_modules_7_modules_branch2_modules_2_modules_bn_parameters_bias_ = (None)
+        x_71 = torch.nn.functional.relu(x_70, inplace=True)
+        x_70 = None
+        input_2 = torch._C._nn.avg_pool2d(out_3, 3, 1, 1, False, False, None)
+        out_3 = None
+        x_72 = torch.conv2d(
+            input_2,
+            l_self_modules_features_modules_7_modules_branch3_modules_1_modules_conv_parameters_weight_,
+            None,
+            (1, 1),
+            (0, 0),
+            (1, 1),
+            1,
+        )
+        input_2 = l_self_modules_features_modules_7_modules_branch3_modules_1_modules_conv_parameters_weight_ = (None)
+        x_73 = torch.nn.functional.batch_norm(
+            x_72,
+            l_self_modules_features_modules_7_modules_branch3_modules_1_modules_bn_buffers_running_mean_,
+            l_self_modules_features_modules_7_modules_branch3_modules_1_modules_bn_buffers_running_var_,
+            l_self_modules_features_modules_7_modules_branch3_modules_1_modules_bn_parameters_weight_,
+            l_self_modules_features_modules_7_modules_branch3_modules_1_modules_bn_parameters_bias_,
+            False,
+            0.1,
+            0.001,
+        )
+        x_72 = l_self_modules_features_modules_7_modules_branch3_modules_1_modules_bn_buffers_running_mean_ = l_self_modules_features_modules_7_modules_branch3_modules_1_modules_bn_buffers_running_var_ = l_self_modules_features_modules_7_modules_branch3_modules_1_modules_bn_parameters_weight_ = l_self_modules_features_modules_7_modules_branch3_modules_1_modules_bn_parameters_bias_ = (None)
+        x_74 = torch.nn.functional.relu(x_73, inplace=True)
+        x_73 = None
+        out_4 = torch.cat((x_56, x_62, x_71, x_74), 1)
+        x_56 = x_62 = x_71 = x_74 = None
+        x_75 = torch.conv2d(
+            out_4,
+            l_self_modules_features_modules_8_modules_branch0_modules_conv_parameters_weight_,
+            None,
+            (1, 1),
+            (0, 0),
+            (1, 1),
+            1,
+        )
+        l_self_modules_features_modules_8_modules_branch0_modules_conv_parameters_weight_ = (
+            None
+        )
+        x_76 = torch.nn.functional.batch_norm(
+            x_75,
+            l_self_modules_features_modules_8_modules_branch0_modules_bn_buffers_running_mean_,
+            l_self_modules_features_modules_8_modules_branch0_modules_bn_buffers_running_var_,
+            l_self_modules_features_modules_8_modules_branch0_modules_bn_parameters_weight_,
+            l_self_modules_features_modules_8_modules_branch0_modules_bn_parameters_bias_,
+            False,
+            0.1,
+            0.001,
+        )
+        x_75 = l_self_modules_features_modules_8_modules_branch0_modules_bn_buffers_running_mean_ = l_self_modules_features_modules_8_modules_branch0_modules_bn_buffers_running_var_ = l_self_modules_features_modules_8_modules_branch0_modules_bn_parameters_weight_ = l_self_modules_features_modules_8_modules_branch0_modules_bn_parameters_bias_ = (None)
+        x_77 = torch.nn.functional.relu(x_76, inplace=True)
+        x_76 = None
+        x_78 = torch.conv2d(
+            out_4,
+            l_self_modules_features_modules_8_modules_branch1_modules_0_modules_conv_parameters_weight_,
+            None,
+            (1, 1),
+            (0, 0),
+            (1, 1),
+            1,
+        )
+        l_self_modules_features_modules_8_modules_branch1_modules_0_modules_conv_parameters_weight_ = (
+            None
+        )
+        x_79 = torch.nn.functional.batch_norm(
+            x_78,
+            l_self_modules_features_modules_8_modules_branch1_modules_0_modules_bn_buffers_running_mean_,
+            l_self_modules_features_modules_8_modules_branch1_modules_0_modules_bn_buffers_running_var_,
+            l_self_modules_features_modules_8_modules_branch1_modules_0_modules_bn_parameters_weight_,
+            l_self_modules_features_modules_8_modules_branch1_modules_0_modules_bn_parameters_bias_,
+            False,
+            0.1,
+            0.001,
+        )
+        x_78 = l_self_modules_features_modules_8_modules_branch1_modules_0_modules_bn_buffers_running_mean_ = l_self_modules_features_modules_8_modules_branch1_modules_0_modules_bn_buffers_running_var_ = l_self_modules_features_modules_8_modules_branch1_modules_0_modules_bn_parameters_weight_ = l_self_modules_features_modules_8_modules_branch1_modules_0_modules_bn_parameters_bias_ = (None)
+        x_80 = torch.nn.functional.relu(x_79, inplace=True)
+        x_79 = None
+        x_81 = torch.conv2d(
+            x_80,
+            l_self_modules_features_modules_8_modules_branch1_modules_1_modules_conv_parameters_weight_,
+            None,
+            (1, 1),
+            (1, 1),
+            (1, 1),
+            1,
+        )
+        x_80 = l_self_modules_features_modules_8_modules_branch1_modules_1_modules_conv_parameters_weight_ = (None)
+        x_82 = torch.nn.functional.batch_norm(
+            x_81,
+            l_self_modules_features_modules_8_modules_branch1_modules_1_modules_bn_buffers_running_mean_,
+            l_self_modules_features_modules_8_modules_branch1_modules_1_modules_bn_buffers_running_var_,
+            l_self_modules_features_modules_8_modules_branch1_modules_1_modules_bn_parameters_weight_,
+            l_self_modules_features_modules_8_modules_branch1_modules_1_modules_bn_parameters_bias_,
+            False,
+            0.1,
+            0.001,
+        )
+        x_81 = l_self_modules_features_modules_8_modules_branch1_modules_1_modules_bn_buffers_running_mean_ = l_self_modules_features_modules_8_modules_branch1_modules_1_modules_bn_buffers_running_var_ = l_self_modules_features_modules_8_modules_branch1_modules_1_modules_bn_parameters_weight_ = l_self_modules_features_modules_8_modules_branch1_modules_1_modules_bn_parameters_bias_ = (None)
+        x_83 = torch.nn.functional.relu(x_82, inplace=True)
+        x_82 = None
+        x_84 = torch.conv2d(
+            out_4,
+            l_self_modules_features_modules_8_modules_branch2_modules_0_modules_conv_parameters_weight_,
+            None,
+            (1, 1),
+            (0, 0),
+            (1, 1),
+            1,
+        )
+        l_self_modules_features_modules_8_modules_branch2_modules_0_modules_conv_parameters_weight_ = (
+            None
+        )
+        x_85 = torch.nn.functional.batch_norm(
+            x_84,
+            l_self_modules_features_modules_8_modules_branch2_modules_0_modules_bn_buffers_running_mean_,
+            l_self_modules_features_modules_8_modules_branch2_modules_0_modules_bn_buffers_running_var_,
+            l_self_modules_features_modules_8_modules_branch2_modules_0_modules_bn_parameters_weight_,
+            l_self_modules_features_modules_8_modules_branch2_modules_0_modules_bn_parameters_bias_,
+            False,
+            0.1,
+            0.001,
+        )
+        x_84 = l_self_modules_features_modules_8_modules_branch2_modules_0_modules_bn_buffers_running_mean_ = l_self_modules_features_modules_8_modules_branch2_modules_0_modules_bn_buffers_running_var_ = l_self_modules_features_modules_8_modules_branch2_modules_0_modules_bn_parameters_weight_ = l_self_modules_features_modules_8_modules_branch2_modules_0_modules_bn_parameters_bias_ = (None)
+        x_86 = torch.nn.functional.relu(x_85, inplace=True)
+        x_85 = None
+        x_87 = torch.conv2d(
+            x_86,
+            l_self_modules_features_modules_8_modules_branch2_modules_1_modules_conv_parameters_weight_,
+            None,
+            (1, 1),
+            (1, 1),
+            (1, 1),
+            1,
+        )
+        x_86 = l_self_modules_features_modules_8_modules_branch2_modules_1_modules_conv_parameters_weight_ = (None)
+        x_88 = torch.nn.functional.batch_norm(
+            x_87,
+            l_self_modules_features_modules_8_modules_branch2_modules_1_modules_bn_buffers_running_mean_,
+            l_self_modules_features_modules_8_modules_branch2_modules_1_modules_bn_buffers_running_var_,
+            l_self_modules_features_modules_8_modules_branch2_modules_1_modules_bn_parameters_weight_,
+            l_self_modules_features_modules_8_modules_branch2_modules_1_modules_bn_parameters_bias_,
+            False,
+            0.1,
+            0.001,
+        )
+        x_87 = l_self_modules_features_modules_8_modules_branch2_modules_1_modules_bn_buffers_running_mean_ = l_self_modules_features_modules_8_modules_branch2_modules_1_modules_bn_buffers_running_var_ = l_self_modules_features_modules_8_modules_branch2_modules_1_modules_bn_parameters_weight_ = l_self_modules_features_modules_8_modules_branch2_modules_1_modules_bn_parameters_bias_ = (None)
+        x_89 = torch.nn.functional.relu(x_88, inplace=True)
+        x_88 = None
+        x_90 = torch.conv2d(
+            x_89,
+            l_self_modules_features_modules_8_modules_branch2_modules_2_modules_conv_parameters_weight_,
+            None,
+            (1, 1),
+            (1, 1),
+            (1, 1),
+            1,
+        )
+        x_89 = l_self_modules_features_modules_8_modules_branch2_modules_2_modules_conv_parameters_weight_ = (None)
+        x_91 = torch.nn.functional.batch_norm(
+            x_90,
+            l_self_modules_features_modules_8_modules_branch2_modules_2_modules_bn_buffers_running_mean_,
+            l_self_modules_features_modules_8_modules_branch2_modules_2_modules_bn_buffers_running_var_,
+            l_self_modules_features_modules_8_modules_branch2_modules_2_modules_bn_parameters_weight_,
+            l_self_modules_features_modules_8_modules_branch2_modules_2_modules_bn_parameters_bias_,
+            False,
+            0.1,
+            0.001,
+        )
+        x_90 = l_self_modules_features_modules_8_modules_branch2_modules_2_modules_bn_buffers_running_mean_ = l_self_modules_features_modules_8_modules_branch2_modules_2_modules_bn_buffers_running_var_ = l_self_modules_features_modules_8_modules_branch2_modules_2_modules_bn_parameters_weight_ = l_self_modules_features_modules_8_modules_branch2_modules_2_modules_bn_parameters_bias_ = (None)
+        x_92 = torch.nn.functional.relu(x_91, inplace=True)
+        x_91 = None
+        input_3 = torch._C._nn.avg_pool2d(out_4, 3, 1, 1, False, False, None)
+        out_4 = None
+        x_93 = torch.conv2d(
+            input_3,
+            l_self_modules_features_modules_8_modules_branch3_modules_1_modules_conv_parameters_weight_,
+            None,
+            (1, 1),
+            (0, 0),
+            (1, 1),
+            1,
+        )
+        input_3 = l_self_modules_features_modules_8_modules_branch3_modules_1_modules_conv_parameters_weight_ = (None)
+        x_94 = torch.nn.functional.batch_norm(
+            x_93,
+            l_self_modules_features_modules_8_modules_branch3_modules_1_modules_bn_buffers_running_mean_,
+            l_self_modules_features_modules_8_modules_branch3_modules_1_modules_bn_buffers_running_var_,
+            l_self_modules_features_modules_8_modules_branch3_modules_1_modules_bn_parameters_weight_,
+            l_self_modules_features_modules_8_modules_branch3_modules_1_modules_bn_parameters_bias_,
+            False,
+            0.1,
+            0.001,
+        )
+        x_93 = l_self_modules_features_modules_8_modules_branch3_modules_1_modules_bn_buffers_running_mean_ = l_self_modules_features_modules_8_modules_branch3_modules_1_modules_bn_buffers_running_var_ = l_self_modules_features_modules_8_modules_branch3_modules_1_modules_bn_parameters_weight_ = l_self_modules_features_modules_8_modules_branch3_modules_1_modules_bn_parameters_bias_ = (None)
+        x_95 = torch.nn.functional.relu(x_94, inplace=True)
+        x_94 = None
+        out_5 = torch.cat((x_77, x_83, x_92, x_95), 1)
+        x_77 = x_83 = x_92 = x_95 = None
+        x_96 = torch.conv2d(
+            out_5,
+            l_self_modules_features_modules_9_modules_branch0_modules_conv_parameters_weight_,
+            None,
+            (1, 1),
+            (0, 0),
+            (1, 1),
+            1,
+        )
+        l_self_modules_features_modules_9_modules_branch0_modules_conv_parameters_weight_ = (
+            None
+        )
+        x_97 = torch.nn.functional.batch_norm(
+            x_96,
+            l_self_modules_features_modules_9_modules_branch0_modules_bn_buffers_running_mean_,
+            l_self_modules_features_modules_9_modules_branch0_modules_bn_buffers_running_var_,
+            l_self_modules_features_modules_9_modules_branch0_modules_bn_parameters_weight_,
+            l_self_modules_features_modules_9_modules_branch0_modules_bn_parameters_bias_,
+            False,
+            0.1,
+            0.001,
+        )
+        x_96 = l_self_modules_features_modules_9_modules_branch0_modules_bn_buffers_running_mean_ = l_self_modules_features_modules_9_modules_branch0_modules_bn_buffers_running_var_ = l_self_modules_features_modules_9_modules_branch0_modules_bn_parameters_weight_ = l_self_modules_features_modules_9_modules_branch0_modules_bn_parameters_bias_ = (None)
+        x_98 = torch.nn.functional.relu(x_97, inplace=True)
+        x_97 = None
+        x_99 = torch.conv2d(
+            out_5,
+            l_self_modules_features_modules_9_modules_branch1_modules_0_modules_conv_parameters_weight_,
+            None,
+            (1, 1),
+            (0, 0),
+            (1, 1),
+            1,
+        )
+        l_self_modules_features_modules_9_modules_branch1_modules_0_modules_conv_parameters_weight_ = (
+            None
+        )
+        x_100 = torch.nn.functional.batch_norm(
+            x_99,
+            l_self_modules_features_modules_9_modules_branch1_modules_0_modules_bn_buffers_running_mean_,
+            l_self_modules_features_modules_9_modules_branch1_modules_0_modules_bn_buffers_running_var_,
+            l_self_modules_features_modules_9_modules_branch1_modules_0_modules_bn_parameters_weight_,
+            l_self_modules_features_modules_9_modules_branch1_modules_0_modules_bn_parameters_bias_,
+            False,
+            0.1,
+            0.001,
+        )
+        x_99 = l_self_modules_features_modules_9_modules_branch1_modules_0_modules_bn_buffers_running_mean_ = l_self_modules_features_modules_9_modules_branch1_modules_0_modules_bn_buffers_running_var_ = l_self_modules_features_modules_9_modules_branch1_modules_0_modules_bn_parameters_weight_ = l_self_modules_features_modules_9_modules_branch1_modules_0_modules_bn_parameters_bias_ = (None)
+        x_101 = torch.nn.functional.relu(x_100, inplace=True)
+        x_100 = None
+        x_102 = torch.conv2d(
+            x_101,
+            l_self_modules_features_modules_9_modules_branch1_modules_1_modules_conv_parameters_weight_,
+            None,
+            (1, 1),
+            (1, 1),
+            (1, 1),
+            1,
+        )
+        x_101 = l_self_modules_features_modules_9_modules_branch1_modules_1_modules_conv_parameters_weight_ = (None)
+        x_103 = torch.nn.functional.batch_norm(
+            x_102,
+            l_self_modules_features_modules_9_modules_branch1_modules_1_modules_bn_buffers_running_mean_,
+            l_self_modules_features_modules_9_modules_branch1_modules_1_modules_bn_buffers_running_var_,
+            l_self_modules_features_modules_9_modules_branch1_modules_1_modules_bn_parameters_weight_,
+            l_self_modules_features_modules_9_modules_branch1_modules_1_modules_bn_parameters_bias_,
+            False,
+            0.1,
+            0.001,
+        )
+        x_102 = l_self_modules_features_modules_9_modules_branch1_modules_1_modules_bn_buffers_running_mean_ = l_self_modules_features_modules_9_modules_branch1_modules_1_modules_bn_buffers_running_var_ = l_self_modules_features_modules_9_modules_branch1_modules_1_modules_bn_parameters_weight_ = l_self_modules_features_modules_9_modules_branch1_modules_1_modules_bn_parameters_bias_ = (None)
+        x_104 = torch.nn.functional.relu(x_103, inplace=True)
+        x_103 = None
+        x_105 = torch.conv2d(
+            out_5,
+            l_self_modules_features_modules_9_modules_branch2_modules_0_modules_conv_parameters_weight_,
+            None,
+            (1, 1),
+            (0, 0),
+            (1, 1),
+            1,
+        )
+        l_self_modules_features_modules_9_modules_branch2_modules_0_modules_conv_parameters_weight_ = (
+            None
+        )
+        x_106 = torch.nn.functional.batch_norm(
+            x_105,
+            l_self_modules_features_modules_9_modules_branch2_modules_0_modules_bn_buffers_running_mean_,
+            l_self_modules_features_modules_9_modules_branch2_modules_0_modules_bn_buffers_running_var_,
+            l_self_modules_features_modules_9_modules_branch2_modules_0_modules_bn_parameters_weight_,
+            l_self_modules_features_modules_9_modules_branch2_modules_0_modules_bn_parameters_bias_,
+            False,
+            0.1,
+            0.001,
+        )
+        x_105 = l_self_modules_features_modules_9_modules_branch2_modules_0_modules_bn_buffers_running_mean_ = l_self_modules_features_modules_9_modules_branch2_modules_0_modules_bn_buffers_running_var_ = l_self_modules_features_modules_9_modules_branch2_modules_0_modules_bn_parameters_weight_ = l_self_modules_features_modules_9_modules_branch2_modules_0_modules_bn_parameters_bias_ = (None)
+        x_107 = torch.nn.functional.relu(x_106, inplace=True)
+        x_106 = None
+        x_108 = torch.conv2d(
+            x_107,
+            l_self_modules_features_modules_9_modules_branch2_modules_1_modules_conv_parameters_weight_,
+            None,
+            (1, 1),
+            (1, 1),
+            (1, 1),
+            1,
+        )
+        x_107 = l_self_modules_features_modules_9_modules_branch2_modules_1_modules_conv_parameters_weight_ = (None)
+        x_109 = torch.nn.functional.batch_norm(
+            x_108,
+            l_self_modules_features_modules_9_modules_branch2_modules_1_modules_bn_buffers_running_mean_,
+            l_self_modules_features_modules_9_modules_branch2_modules_1_modules_bn_buffers_running_var_,
+            l_self_modules_features_modules_9_modules_branch2_modules_1_modules_bn_parameters_weight_,
+            l_self_modules_features_modules_9_modules_branch2_modules_1_modules_bn_parameters_bias_,
+            False,
+            0.1,
+            0.001,
+        )
+        x_108 = l_self_modules_features_modules_9_modules_branch2_modules_1_modules_bn_buffers_running_mean_ = l_self_modules_features_modules_9_modules_branch2_modules_1_modules_bn_buffers_running_var_ = l_self_modules_features_modules_9_modules_branch2_modules_1_modules_bn_parameters_weight_ = l_self_modules_features_modules_9_modules_branch2_modules_1_modules_bn_parameters_bias_ = (None)
+        x_110 = torch.nn.functional.relu(x_109, inplace=True)
+        x_109 = None
+        x_111 = torch.conv2d(
+            x_110,
+            l_self_modules_features_modules_9_modules_branch2_modules_2_modules_conv_parameters_weight_,
+            None,
+            (1, 1),
+            (1, 1),
+            (1, 1),
+            1,
+        )
+        x_110 = l_self_modules_features_modules_9_modules_branch2_modules_2_modules_conv_parameters_weight_ = (None)
+        x_112 = torch.nn.functional.batch_norm(
+            x_111,
+            l_self_modules_features_modules_9_modules_branch2_modules_2_modules_bn_buffers_running_mean_,
+            l_self_modules_features_modules_9_modules_branch2_modules_2_modules_bn_buffers_running_var_,
+            l_self_modules_features_modules_9_modules_branch2_modules_2_modules_bn_parameters_weight_,
+            l_self_modules_features_modules_9_modules_branch2_modules_2_modules_bn_parameters_bias_,
+            False,
+            0.1,
+            0.001,
+        )
+        x_111 = l_self_modules_features_modules_9_modules_branch2_modules_2_modules_bn_buffers_running_mean_ = l_self_modules_features_modules_9_modules_branch2_modules_2_modules_bn_buffers_running_var_ = l_self_modules_features_modules_9_modules_branch2_modules_2_modules_bn_parameters_weight_ = l_self_modules_features_modules_9_modules_branch2_modules_2_modules_bn_parameters_bias_ = (None)
+        x_113 = torch.nn.functional.relu(x_112, inplace=True)
+        x_112 = None
+        input_4 = torch._C._nn.avg_pool2d(out_5, 3, 1, 1, False, False, None)
+        out_5 = None
+        x_114 = torch.conv2d(
+            input_4,
+            l_self_modules_features_modules_9_modules_branch3_modules_1_modules_conv_parameters_weight_,
+            None,
+            (1, 1),
+            (0, 0),
+            (1, 1),
+            1,
+        )
+        input_4 = l_self_modules_features_modules_9_modules_branch3_modules_1_modules_conv_parameters_weight_ = (None)
+        x_115 = torch.nn.functional.batch_norm(
+            x_114,
+            l_self_modules_features_modules_9_modules_branch3_modules_1_modules_bn_buffers_running_mean_,
+            l_self_modules_features_modules_9_modules_branch3_modules_1_modules_bn_buffers_running_var_,
+            l_self_modules_features_modules_9_modules_branch3_modules_1_modules_bn_parameters_weight_,
+            l_self_modules_features_modules_9_modules_branch3_modules_1_modules_bn_parameters_bias_,
+            False,
+            0.1,
+            0.001,
+        )
+        x_114 = l_self_modules_features_modules_9_modules_branch3_modules_1_modules_bn_buffers_running_mean_ = l_self_modules_features_modules_9_modules_branch3_modules_1_modules_bn_buffers_running_var_ = l_self_modules_features_modules_9_modules_branch3_modules_1_modules_bn_parameters_weight_ = l_self_modules_features_modules_9_modules_branch3_modules_1_modules_bn_parameters_bias_ = (None)
+        x_116 = torch.nn.functional.relu(x_115, inplace=True)
+        x_115 = None
+        out_6 = torch.cat((x_98, x_104, x_113, x_116), 1)
+        x_98 = x_104 = x_113 = x_116 = None
+        x_117 = torch.conv2d(
+            out_6,
+            l_self_modules_features_modules_10_modules_branch0_modules_conv_parameters_weight_,
+            None,
+            (2, 2),
+            (0, 0),
+            (1, 1),
+            1,
+        )
+        l_self_modules_features_modules_10_modules_branch0_modules_conv_parameters_weight_ = (
+            None
+        )
+        x_118 = torch.nn.functional.batch_norm(
+            x_117,
+            l_self_modules_features_modules_10_modules_branch0_modules_bn_buffers_running_mean_,
+            l_self_modules_features_modules_10_modules_branch0_modules_bn_buffers_running_var_,
+            l_self_modules_features_modules_10_modules_branch0_modules_bn_parameters_weight_,
+            l_self_modules_features_modules_10_modules_branch0_modules_bn_parameters_bias_,
+            False,
+            0.1,
+            0.001,
+        )
+        x_117 = l_self_modules_features_modules_10_modules_branch0_modules_bn_buffers_running_mean_ = l_self_modules_features_modules_10_modules_branch0_modules_bn_buffers_running_var_ = l_self_modules_features_modules_10_modules_branch0_modules_bn_parameters_weight_ = l_self_modules_features_modules_10_modules_branch0_modules_bn_parameters_bias_ = (None)
+        x_119 = torch.nn.functional.relu(x_118, inplace=True)
+        x_118 = None
+        x_120 = torch.conv2d(
+            out_6,
+            l_self_modules_features_modules_10_modules_branch1_modules_0_modules_conv_parameters_weight_,
+            None,
+            (1, 1),
+            (0, 0),
+            (1, 1),
+            1,
+        )
+        l_self_modules_features_modules_10_modules_branch1_modules_0_modules_conv_parameters_weight_ = (
+            None
+        )
+        x_121 = torch.nn.functional.batch_norm(
+            x_120,
+            l_self_modules_features_modules_10_modules_branch1_modules_0_modules_bn_buffers_running_mean_,
+            l_self_modules_features_modules_10_modules_branch1_modules_0_modules_bn_buffers_running_var_,
+            l_self_modules_features_modules_10_modules_branch1_modules_0_modules_bn_parameters_weight_,
+            l_self_modules_features_modules_10_modules_branch1_modules_0_modules_bn_parameters_bias_,
+            False,
+            0.1,
+            0.001,
+        )
+        x_120 = l_self_modules_features_modules_10_modules_branch1_modules_0_modules_bn_buffers_running_mean_ = l_self_modules_features_modules_10_modules_branch1_modules_0_modules_bn_buffers_running_var_ = l_self_modules_features_modules_10_modules_branch1_modules_0_modules_bn_parameters_weight_ = l_self_modules_features_modules_10_modules_branch1_modules_0_modules_bn_parameters_bias_ = (None)
+        x_122 = torch.nn.functional.relu(x_121, inplace=True)
+        x_121 = None
+        x_123 = torch.conv2d(
+            x_122,
+            l_self_modules_features_modules_10_modules_branch1_modules_1_modules_conv_parameters_weight_,
+            None,
+            (1, 1),
+            (1, 1),
+            (1, 1),
+            1,
+        )
+        x_122 = l_self_modules_features_modules_10_modules_branch1_modules_1_modules_conv_parameters_weight_ = (None)
+        x_124 = torch.nn.functional.batch_norm(
+            x_123,
+            l_self_modules_features_modules_10_modules_branch1_modules_1_modules_bn_buffers_running_mean_,
+            l_self_modules_features_modules_10_modules_branch1_modules_1_modules_bn_buffers_running_var_,
+            l_self_modules_features_modules_10_modules_branch1_modules_1_modules_bn_parameters_weight_,
+            l_self_modules_features_modules_10_modules_branch1_modules_1_modules_bn_parameters_bias_,
+            False,
+            0.1,
+            0.001,
+        )
+        x_123 = l_self_modules_features_modules_10_modules_branch1_modules_1_modules_bn_buffers_running_mean_ = l_self_modules_features_modules_10_modules_branch1_modules_1_modules_bn_buffers_running_var_ = l_self_modules_features_modules_10_modules_branch1_modules_1_modules_bn_parameters_weight_ = l_self_modules_features_modules_10_modules_branch1_modules_1_modules_bn_parameters_bias_ = (None)
+        x_125 = torch.nn.functional.relu(x_124, inplace=True)
+        x_124 = None
+        x_126 = torch.conv2d(
+            x_125,
+            l_self_modules_features_modules_10_modules_branch1_modules_2_modules_conv_parameters_weight_,
+            None,
+            (2, 2),
+            (0, 0),
+            (1, 1),
+            1,
+        )
+        x_125 = l_self_modules_features_modules_10_modules_branch1_modules_2_modules_conv_parameters_weight_ = (None)
+        x_127 = torch.nn.functional.batch_norm(
+            x_126,
+            l_self_modules_features_modules_10_modules_branch1_modules_2_modules_bn_buffers_running_mean_,
+            l_self_modules_features_modules_10_modules_branch1_modules_2_modules_bn_buffers_running_var_,
+            l_self_modules_features_modules_10_modules_branch1_modules_2_modules_bn_parameters_weight_,
+            l_self_modules_features_modules_10_modules_branch1_modules_2_modules_bn_parameters_bias_,
+            False,
+            0.1,
+            0.001,
+        )
+        x_126 = l_self_modules_features_modules_10_modules_branch1_modules_2_modules_bn_buffers_running_mean_ = l_self_modules_features_modules_10_modules_branch1_modules_2_modules_bn_buffers_running_var_ = l_self_modules_features_modules_10_modules_branch1_modules_2_modules_bn_parameters_weight_ = l_self_modules_features_modules_10_modules_branch1_modules_2_modules_bn_parameters_bias_ = (None)
+        x_128 = torch.nn.functional.relu(x_127, inplace=True)
+        x_127 = None
+        x2 = torch.nn.functional.max_pool2d(
+            out_6, 3, 2, 0, 1, ceil_mode=False, return_indices=False
+        )
+        out_6 = None
+        out_7 = torch.cat((x_119, x_128, x2), 1)
+        x_119 = x_128 = x2 = None
+        x_129 = torch.conv2d(
+            out_7,
+            l_self_modules_features_modules_11_modules_branch0_modules_conv_parameters_weight_,
+            None,
+            (1, 1),
+            (0, 0),
+            (1, 1),
+            1,
+        )
+        l_self_modules_features_modules_11_modules_branch0_modules_conv_parameters_weight_ = (
+            None
+        )
+        x_130 = torch.nn.functional.batch_norm(
+            x_129,
+            l_self_modules_features_modules_11_modules_branch0_modules_bn_buffers_running_mean_,
+            l_self_modules_features_modules_11_modules_branch0_modules_bn_buffers_running_var_,
+            l_self_modules_features_modules_11_modules_branch0_modules_bn_parameters_weight_,
+            l_self_modules_features_modules_11_modules_branch0_modules_bn_parameters_bias_,
+            False,
+            0.1,
+            0.001,
+        )
+        x_129 = l_self_modules_features_modules_11_modules_branch0_modules_bn_buffers_running_mean_ = l_self_modules_features_modules_11_modules_branch0_modules_bn_buffers_running_var_ = l_self_modules_features_modules_11_modules_branch0_modules_bn_parameters_weight_ = l_self_modules_features_modules_11_modules_branch0_modules_bn_parameters_bias_ = (None)
+        x_131 = torch.nn.functional.relu(x_130, inplace=True)
+        x_130 = None
+        x_132 = torch.conv2d(
+            out_7,
+            l_self_modules_features_modules_11_modules_branch1_modules_0_modules_conv_parameters_weight_,
+            None,
+            (1, 1),
+            (0, 0),
+            (1, 1),
+            1,
+        )
+        l_self_modules_features_modules_11_modules_branch1_modules_0_modules_conv_parameters_weight_ = (
+            None
+        )
+        x_133 = torch.nn.functional.batch_norm(
+            x_132,
+            l_self_modules_features_modules_11_modules_branch1_modules_0_modules_bn_buffers_running_mean_,
+            l_self_modules_features_modules_11_modules_branch1_modules_0_modules_bn_buffers_running_var_,
+            l_self_modules_features_modules_11_modules_branch1_modules_0_modules_bn_parameters_weight_,
+            l_self_modules_features_modules_11_modules_branch1_modules_0_modules_bn_parameters_bias_,
+            False,
+            0.1,
+            0.001,
+        )
+        x_132 = l_self_modules_features_modules_11_modules_branch1_modules_0_modules_bn_buffers_running_mean_ = l_self_modules_features_modules_11_modules_branch1_modules_0_modules_bn_buffers_running_var_ = l_self_modules_features_modules_11_modules_branch1_modules_0_modules_bn_parameters_weight_ = l_self_modules_features_modules_11_modules_branch1_modules_0_modules_bn_parameters_bias_ = (None)
+        x_134 = torch.nn.functional.relu(x_133, inplace=True)
+        x_133 = None
+        x_135 = torch.conv2d(
+            x_134,
+            l_self_modules_features_modules_11_modules_branch1_modules_1_modules_conv_parameters_weight_,
+            None,
+            (1, 1),
+            (0, 3),
+            (1, 1),
+            1,
+        )
+        x_134 = l_self_modules_features_modules_11_modules_branch1_modules_1_modules_conv_parameters_weight_ = (None)
+        x_136 = torch.nn.functional.batch_norm(
+            x_135,
+            l_self_modules_features_modules_11_modules_branch1_modules_1_modules_bn_buffers_running_mean_,
+            l_self_modules_features_modules_11_modules_branch1_modules_1_modules_bn_buffers_running_var_,
+            l_self_modules_features_modules_11_modules_branch1_modules_1_modules_bn_parameters_weight_,
+            l_self_modules_features_modules_11_modules_branch1_modules_1_modules_bn_parameters_bias_,
+            False,
+            0.1,
+            0.001,
+        )
+        x_135 = l_self_modules_features_modules_11_modules_branch1_modules_1_modules_bn_buffers_running_mean_ = l_self_modules_features_modules_11_modules_branch1_modules_1_modules_bn_buffers_running_var_ = l_self_modules_features_modules_11_modules_branch1_modules_1_modules_bn_parameters_weight_ = l_self_modules_features_modules_11_modules_branch1_modules_1_modules_bn_parameters_bias_ = (None)
+        x_137 = torch.nn.functional.relu(x_136, inplace=True)
+        x_136 = None
+        x_138 = torch.conv2d(
+            x_137,
+            l_self_modules_features_modules_11_modules_branch1_modules_2_modules_conv_parameters_weight_,
+            None,
+            (1, 1),
+            (3, 0),
+            (1, 1),
+            1,
+        )
+        x_137 = l_self_modules_features_modules_11_modules_branch1_modules_2_modules_conv_parameters_weight_ = (None)
+        x_139 = torch.nn.functional.batch_norm(
+            x_138,
+            l_self_modules_features_modules_11_modules_branch1_modules_2_modules_bn_buffers_running_mean_,
+            l_self_modules_features_modules_11_modules_branch1_modules_2_modules_bn_buffers_running_var_,
+            l_self_modules_features_modules_11_modules_branch1_modules_2_modules_bn_parameters_weight_,
+            l_self_modules_features_modules_11_modules_branch1_modules_2_modules_bn_parameters_bias_,
+            False,
+            0.1,
+            0.001,
+        )
+        x_138 = l_self_modules_features_modules_11_modules_branch1_modules_2_modules_bn_buffers_running_mean_ = l_self_modules_features_modules_11_modules_branch1_modules_2_modules_bn_buffers_running_var_ = l_self_modules_features_modules_11_modules_branch1_modules_2_modules_bn_parameters_weight_ = l_self_modules_features_modules_11_modules_branch1_modules_2_modules_bn_parameters_bias_ = (None)
+        x_140 = torch.nn.functional.relu(x_139, inplace=True)
+        x_139 = None
+        x_141 = torch.conv2d(
+            out_7,
+            l_self_modules_features_modules_11_modules_branch2_modules_0_modules_conv_parameters_weight_,
+            None,
+            (1, 1),
+            (0, 0),
+            (1, 1),
+            1,
+        )
+        l_self_modules_features_modules_11_modules_branch2_modules_0_modules_conv_parameters_weight_ = (
+            None
+        )
+        x_142 = torch.nn.functional.batch_norm(
+            x_141,
+            l_self_modules_features_modules_11_modules_branch2_modules_0_modules_bn_buffers_running_mean_,
+            l_self_modules_features_modules_11_modules_branch2_modules_0_modules_bn_buffers_running_var_,
+            l_self_modules_features_modules_11_modules_branch2_modules_0_modules_bn_parameters_weight_,
+            l_self_modules_features_modules_11_modules_branch2_modules_0_modules_bn_parameters_bias_,
+            False,
+            0.1,
+            0.001,
+        )
+        x_141 = l_self_modules_features_modules_11_modules_branch2_modules_0_modules_bn_buffers_running_mean_ = l_self_modules_features_modules_11_modules_branch2_modules_0_modules_bn_buffers_running_var_ = l_self_modules_features_modules_11_modules_branch2_modules_0_modules_bn_parameters_weight_ = l_self_modules_features_modules_11_modules_branch2_modules_0_modules_bn_parameters_bias_ = (None)
+        x_143 = torch.nn.functional.relu(x_142, inplace=True)
+        x_142 = None
+        x_144 = torch.conv2d(
+            x_143,
+            l_self_modules_features_modules_11_modules_branch2_modules_1_modules_conv_parameters_weight_,
+            None,
+            (1, 1),
+            (3, 0),
+            (1, 1),
+            1,
+        )
+        x_143 = l_self_modules_features_modules_11_modules_branch2_modules_1_modules_conv_parameters_weight_ = (None)
+        x_145 = torch.nn.functional.batch_norm(
+            x_144,
+            l_self_modules_features_modules_11_modules_branch2_modules_1_modules_bn_buffers_running_mean_,
+            l_self_modules_features_modules_11_modules_branch2_modules_1_modules_bn_buffers_running_var_,
+            l_self_modules_features_modules_11_modules_branch2_modules_1_modules_bn_parameters_weight_,
+            l_self_modules_features_modules_11_modules_branch2_modules_1_modules_bn_parameters_bias_,
+            False,
+            0.1,
+            0.001,
+        )
+        x_144 = l_self_modules_features_modules_11_modules_branch2_modules_1_modules_bn_buffers_running_mean_ = l_self_modules_features_modules_11_modules_branch2_modules_1_modules_bn_buffers_running_var_ = l_self_modules_features_modules_11_modules_branch2_modules_1_modules_bn_parameters_weight_ = l_self_modules_features_modules_11_modules_branch2_modules_1_modules_bn_parameters_bias_ = (None)
+        x_146 = torch.nn.functional.relu(x_145, inplace=True)
+        x_145 = None
+        x_147 = torch.conv2d(
+            x_146,
+            l_self_modules_features_modules_11_modules_branch2_modules_2_modules_conv_parameters_weight_,
+            None,
+            (1, 1),
+            (0, 3),
+            (1, 1),
+            1,
+        )
+        x_146 = l_self_modules_features_modules_11_modules_branch2_modules_2_modules_conv_parameters_weight_ = (None)
+        x_148 = torch.nn.functional.batch_norm(
+            x_147,
+            l_self_modules_features_modules_11_modules_branch2_modules_2_modules_bn_buffers_running_mean_,
+            l_self_modules_features_modules_11_modules_branch2_modules_2_modules_bn_buffers_running_var_,
+            l_self_modules_features_modules_11_modules_branch2_modules_2_modules_bn_parameters_weight_,
+            l_self_modules_features_modules_11_modules_branch2_modules_2_modules_bn_parameters_bias_,
+            False,
+            0.1,
+            0.001,
+        )
+        x_147 = l_self_modules_features_modules_11_modules_branch2_modules_2_modules_bn_buffers_running_mean_ = l_self_modules_features_modules_11_modules_branch2_modules_2_modules_bn_buffers_running_var_ = l_self_modules_features_modules_11_modules_branch2_modules_2_modules_bn_parameters_weight_ = l_self_modules_features_modules_11_modules_branch2_modules_2_modules_bn_parameters_bias_ = (None)
+        x_149 = torch.nn.functional.relu(x_148, inplace=True)
+        x_148 = None
+        x_150 = torch.conv2d(
+            x_149,
+            l_self_modules_features_modules_11_modules_branch2_modules_3_modules_conv_parameters_weight_,
+            None,
+            (1, 1),
+            (3, 0),
+            (1, 1),
+            1,
+        )
+        x_149 = l_self_modules_features_modules_11_modules_branch2_modules_3_modules_conv_parameters_weight_ = (None)
+        x_151 = torch.nn.functional.batch_norm(
+            x_150,
+            l_self_modules_features_modules_11_modules_branch2_modules_3_modules_bn_buffers_running_mean_,
+            l_self_modules_features_modules_11_modules_branch2_modules_3_modules_bn_buffers_running_var_,
+            l_self_modules_features_modules_11_modules_branch2_modules_3_modules_bn_parameters_weight_,
+            l_self_modules_features_modules_11_modules_branch2_modules_3_modules_bn_parameters_bias_,
+            False,
+            0.1,
+            0.001,
+        )
+        x_150 = l_self_modules_features_modules_11_modules_branch2_modules_3_modules_bn_buffers_running_mean_ = l_self_modules_features_modules_11_modules_branch2_modules_3_modules_bn_buffers_running_var_ = l_self_modules_features_modules_11_modules_branch2_modules_3_modules_bn_parameters_weight_ = l_self_modules_features_modules_11_modules_branch2_modules_3_modules_bn_parameters_bias_ = (None)
+        x_152 = torch.nn.functional.relu(x_151, inplace=True)
+        x_151 = None
+        x_153 = torch.conv2d(
+            x_152,
+            l_self_modules_features_modules_11_modules_branch2_modules_4_modules_conv_parameters_weight_,
+            None,
+            (1, 1),
+            (0, 3),
+            (1, 1),
+            1,
+        )
+        x_152 = l_self_modules_features_modules_11_modules_branch2_modules_4_modules_conv_parameters_weight_ = (None)
+        x_154 = torch.nn.functional.batch_norm(
+            x_153,
+            l_self_modules_features_modules_11_modules_branch2_modules_4_modules_bn_buffers_running_mean_,
+            l_self_modules_features_modules_11_modules_branch2_modules_4_modules_bn_buffers_running_var_,
+            l_self_modules_features_modules_11_modules_branch2_modules_4_modules_bn_parameters_weight_,
+            l_self_modules_features_modules_11_modules_branch2_modules_4_modules_bn_parameters_bias_,
+            False,
+            0.1,
+            0.001,
+        )
+        x_153 = l_self_modules_features_modules_11_modules_branch2_modules_4_modules_bn_buffers_running_mean_ = l_self_modules_features_modules_11_modules_branch2_modules_4_modules_bn_buffers_running_var_ = l_self_modules_features_modules_11_modules_branch2_modules_4_modules_bn_parameters_weight_ = l_self_modules_features_modules_11_modules_branch2_modules_4_modules_bn_parameters_bias_ = (None)
+        x_155 = torch.nn.functional.relu(x_154, inplace=True)
+        x_154 = None
+        input_5 = torch._C._nn.avg_pool2d(out_7, 3, 1, 1, False, False, None)
+        out_7 = None
+        x_156 = torch.conv2d(
+            input_5,
+            l_self_modules_features_modules_11_modules_branch3_modules_1_modules_conv_parameters_weight_,
+            None,
+            (1, 1),
+            (0, 0),
+            (1, 1),
+            1,
+        )
+        input_5 = l_self_modules_features_modules_11_modules_branch3_modules_1_modules_conv_parameters_weight_ = (None)
+        x_157 = torch.nn.functional.batch_norm(
+            x_156,
+            l_self_modules_features_modules_11_modules_branch3_modules_1_modules_bn_buffers_running_mean_,
+            l_self_modules_features_modules_11_modules_branch3_modules_1_modules_bn_buffers_running_var_,
+            l_self_modules_features_modules_11_modules_branch3_modules_1_modules_bn_parameters_weight_,
+            l_self_modules_features_modules_11_modules_branch3_modules_1_modules_bn_parameters_bias_,
+            False,
+            0.1,
+            0.001,
+        )
+        x_156 = l_self_modules_features_modules_11_modules_branch3_modules_1_modules_bn_buffers_running_mean_ = l_self_modules_features_modules_11_modules_branch3_modules_1_modules_bn_buffers_running_var_ = l_self_modules_features_modules_11_modules_branch3_modules_1_modules_bn_parameters_weight_ = l_self_modules_features_modules_11_modules_branch3_modules_1_modules_bn_parameters_bias_ = (None)
+        x_158 = torch.nn.functional.relu(x_157, inplace=True)
+        x_157 = None
+        out_8 = torch.cat((x_131, x_140, x_155, x_158), 1)
+        x_131 = x_140 = x_155 = x_158 = None
+        x_159 = torch.conv2d(
+            out_8,
+            l_self_modules_features_modules_12_modules_branch0_modules_conv_parameters_weight_,
+            None,
+            (1, 1),
+            (0, 0),
+            (1, 1),
+            1,
+        )
+        l_self_modules_features_modules_12_modules_branch0_modules_conv_parameters_weight_ = (
+            None
+        )
+        x_160 = torch.nn.functional.batch_norm(
+            x_159,
+            l_self_modules_features_modules_12_modules_branch0_modules_bn_buffers_running_mean_,
+            l_self_modules_features_modules_12_modules_branch0_modules_bn_buffers_running_var_,
+            l_self_modules_features_modules_12_modules_branch0_modules_bn_parameters_weight_,
+            l_self_modules_features_modules_12_modules_branch0_modules_bn_parameters_bias_,
+            False,
+            0.1,
+            0.001,
+        )
+        x_159 = l_self_modules_features_modules_12_modules_branch0_modules_bn_buffers_running_mean_ = l_self_modules_features_modules_12_modules_branch0_modules_bn_buffers_running_var_ = l_self_modules_features_modules_12_modules_branch0_modules_bn_parameters_weight_ = l_self_modules_features_modules_12_modules_branch0_modules_bn_parameters_bias_ = (None)
+        x_161 = torch.nn.functional.relu(x_160, inplace=True)
+        x_160 = None
+        x_162 = torch.conv2d(
+            out_8,
+            l_self_modules_features_modules_12_modules_branch1_modules_0_modules_conv_parameters_weight_,
+            None,
+            (1, 1),
+            (0, 0),
+            (1, 1),
+            1,
+        )
+        l_self_modules_features_modules_12_modules_branch1_modules_0_modules_conv_parameters_weight_ = (
+            None
+        )
+        x_163 = torch.nn.functional.batch_norm(
+            x_162,
+            l_self_modules_features_modules_12_modules_branch1_modules_0_modules_bn_buffers_running_mean_,
+            l_self_modules_features_modules_12_modules_branch1_modules_0_modules_bn_buffers_running_var_,
+            l_self_modules_features_modules_12_modules_branch1_modules_0_modules_bn_parameters_weight_,
+            l_self_modules_features_modules_12_modules_branch1_modules_0_modules_bn_parameters_bias_,
+            False,
+            0.1,
+            0.001,
+        )
+        x_162 = l_self_modules_features_modules_12_modules_branch1_modules_0_modules_bn_buffers_running_mean_ = l_self_modules_features_modules_12_modules_branch1_modules_0_modules_bn_buffers_running_var_ = l_self_modules_features_modules_12_modules_branch1_modules_0_modules_bn_parameters_weight_ = l_self_modules_features_modules_12_modules_branch1_modules_0_modules_bn_parameters_bias_ = (None)
+        x_164 = torch.nn.functional.relu(x_163, inplace=True)
+        x_163 = None
+        x_165 = torch.conv2d(
+            x_164,
+            l_self_modules_features_modules_12_modules_branch1_modules_1_modules_conv_parameters_weight_,
+            None,
+            (1, 1),
+            (0, 3),
+            (1, 1),
+            1,
+        )
+        x_164 = l_self_modules_features_modules_12_modules_branch1_modules_1_modules_conv_parameters_weight_ = (None)
+        x_166 = torch.nn.functional.batch_norm(
+            x_165,
+            l_self_modules_features_modules_12_modules_branch1_modules_1_modules_bn_buffers_running_mean_,
+            l_self_modules_features_modules_12_modules_branch1_modules_1_modules_bn_buffers_running_var_,
+            l_self_modules_features_modules_12_modules_branch1_modules_1_modules_bn_parameters_weight_,
+            l_self_modules_features_modules_12_modules_branch1_modules_1_modules_bn_parameters_bias_,
+            False,
+            0.1,
+            0.001,
+        )
+        x_165 = l_self_modules_features_modules_12_modules_branch1_modules_1_modules_bn_buffers_running_mean_ = l_self_modules_features_modules_12_modules_branch1_modules_1_modules_bn_buffers_running_var_ = l_self_modules_features_modules_12_modules_branch1_modules_1_modules_bn_parameters_weight_ = l_self_modules_features_modules_12_modules_branch1_modules_1_modules_bn_parameters_bias_ = (None)
+        x_167 = torch.nn.functional.relu(x_166, inplace=True)
+        x_166 = None
+        x_168 = torch.conv2d(
+            x_167,
+            l_self_modules_features_modules_12_modules_branch1_modules_2_modules_conv_parameters_weight_,
+            None,
+            (1, 1),
+            (3, 0),
+            (1, 1),
+            1,
+        )
+        x_167 = l_self_modules_features_modules_12_modules_branch1_modules_2_modules_conv_parameters_weight_ = (None)
+        x_169 = torch.nn.functional.batch_norm(
+            x_168,
+            l_self_modules_features_modules_12_modules_branch1_modules_2_modules_bn_buffers_running_mean_,
+            l_self_modules_features_modules_12_modules_branch1_modules_2_modules_bn_buffers_running_var_,
+            l_self_modules_features_modules_12_modules_branch1_modules_2_modules_bn_parameters_weight_,
+            l_self_modules_features_modules_12_modules_branch1_modules_2_modules_bn_parameters_bias_,
+            False,
+            0.1,
+            0.001,
+        )
+        x_168 = l_self_modules_features_modules_12_modules_branch1_modules_2_modules_bn_buffers_running_mean_ = l_self_modules_features_modules_12_modules_branch1_modules_2_modules_bn_buffers_running_var_ = l_self_modules_features_modules_12_modules_branch1_modules_2_modules_bn_parameters_weight_ = l_self_modules_features_modules_12_modules_branch1_modules_2_modules_bn_parameters_bias_ = (None)
+        x_170 = torch.nn.functional.relu(x_169, inplace=True)
+        x_169 = None
+        x_171 = torch.conv2d(
+            out_8,
+            l_self_modules_features_modules_12_modules_branch2_modules_0_modules_conv_parameters_weight_,
+            None,
+            (1, 1),
+            (0, 0),
+            (1, 1),
+            1,
+        )
+        l_self_modules_features_modules_12_modules_branch2_modules_0_modules_conv_parameters_weight_ = (
+            None
+        )
+        x_172 = torch.nn.functional.batch_norm(
+            x_171,
+            l_self_modules_features_modules_12_modules_branch2_modules_0_modules_bn_buffers_running_mean_,
+            l_self_modules_features_modules_12_modules_branch2_modules_0_modules_bn_buffers_running_var_,
+            l_self_modules_features_modules_12_modules_branch2_modules_0_modules_bn_parameters_weight_,
+            l_self_modules_features_modules_12_modules_branch2_modules_0_modules_bn_parameters_bias_,
+            False,
+            0.1,
+            0.001,
+        )
+        x_171 = l_self_modules_features_modules_12_modules_branch2_modules_0_modules_bn_buffers_running_mean_ = l_self_modules_features_modules_12_modules_branch2_modules_0_modules_bn_buffers_running_var_ = l_self_modules_features_modules_12_modules_branch2_modules_0_modules_bn_parameters_weight_ = l_self_modules_features_modules_12_modules_branch2_modules_0_modules_bn_parameters_bias_ = (None)
+        x_173 = torch.nn.functional.relu(x_172, inplace=True)
+        x_172 = None
+        x_174 = torch.conv2d(
+            x_173,
+            l_self_modules_features_modules_12_modules_branch2_modules_1_modules_conv_parameters_weight_,
+            None,
+            (1, 1),
+            (3, 0),
+            (1, 1),
+            1,
+        )
+        x_173 = l_self_modules_features_modules_12_modules_branch2_modules_1_modules_conv_parameters_weight_ = (None)
+        x_175 = torch.nn.functional.batch_norm(
+            x_174,
+            l_self_modules_features_modules_12_modules_branch2_modules_1_modules_bn_buffers_running_mean_,
+            l_self_modules_features_modules_12_modules_branch2_modules_1_modules_bn_buffers_running_var_,
+            l_self_modules_features_modules_12_modules_branch2_modules_1_modules_bn_parameters_weight_,
+            l_self_modules_features_modules_12_modules_branch2_modules_1_modules_bn_parameters_bias_,
+            False,
+            0.1,
+            0.001,
+        )
+        x_174 = l_self_modules_features_modules_12_modules_branch2_modules_1_modules_bn_buffers_running_mean_ = l_self_modules_features_modules_12_modules_branch2_modules_1_modules_bn_buffers_running_var_ = l_self_modules_features_modules_12_modules_branch2_modules_1_modules_bn_parameters_weight_ = l_self_modules_features_modules_12_modules_branch2_modules_1_modules_bn_parameters_bias_ = (None)
+        x_176 = torch.nn.functional.relu(x_175, inplace=True)
+        x_175 = None
+        x_177 = torch.conv2d(
+            x_176,
+            l_self_modules_features_modules_12_modules_branch2_modules_2_modules_conv_parameters_weight_,
+            None,
+            (1, 1),
+            (0, 3),
+            (1, 1),
+            1,
+        )
+        x_176 = l_self_modules_features_modules_12_modules_branch2_modules_2_modules_conv_parameters_weight_ = (None)
+        x_178 = torch.nn.functional.batch_norm(
+            x_177,
+            l_self_modules_features_modules_12_modules_branch2_modules_2_modules_bn_buffers_running_mean_,
+            l_self_modules_features_modules_12_modules_branch2_modules_2_modules_bn_buffers_running_var_,
+            l_self_modules_features_modules_12_modules_branch2_modules_2_modules_bn_parameters_weight_,
+            l_self_modules_features_modules_12_modules_branch2_modules_2_modules_bn_parameters_bias_,
+            False,
+            0.1,
+            0.001,
+        )
+        x_177 = l_self_modules_features_modules_12_modules_branch2_modules_2_modules_bn_buffers_running_mean_ = l_self_modules_features_modules_12_modules_branch2_modules_2_modules_bn_buffers_running_var_ = l_self_modules_features_modules_12_modules_branch2_modules_2_modules_bn_parameters_weight_ = l_self_modules_features_modules_12_modules_branch2_modules_2_modules_bn_parameters_bias_ = (None)
+        x_179 = torch.nn.functional.relu(x_178, inplace=True)
+        x_178 = None
+        x_180 = torch.conv2d(
+            x_179,
+            l_self_modules_features_modules_12_modules_branch2_modules_3_modules_conv_parameters_weight_,
+            None,
+            (1, 1),
+            (3, 0),
+            (1, 1),
+            1,
+        )
+        x_179 = l_self_modules_features_modules_12_modules_branch2_modules_3_modules_conv_parameters_weight_ = (None)
+        x_181 = torch.nn.functional.batch_norm(
+            x_180,
+            l_self_modules_features_modules_12_modules_branch2_modules_3_modules_bn_buffers_running_mean_,
+            l_self_modules_features_modules_12_modules_branch2_modules_3_modules_bn_buffers_running_var_,
+            l_self_modules_features_modules_12_modules_branch2_modules_3_modules_bn_parameters_weight_,
+            l_self_modules_features_modules_12_modules_branch2_modules_3_modules_bn_parameters_bias_,
+            False,
+            0.1,
+            0.001,
+        )
+        x_180 = l_self_modules_features_modules_12_modules_branch2_modules_3_modules_bn_buffers_running_mean_ = l_self_modules_features_modules_12_modules_branch2_modules_3_modules_bn_buffers_running_var_ = l_self_modules_features_modules_12_modules_branch2_modules_3_modules_bn_parameters_weight_ = l_self_modules_features_modules_12_modules_branch2_modules_3_modules_bn_parameters_bias_ = (None)
+        x_182 = torch.nn.functional.relu(x_181, inplace=True)
+        x_181 = None
+        x_183 = torch.conv2d(
+            x_182,
+            l_self_modules_features_modules_12_modules_branch2_modules_4_modules_conv_parameters_weight_,
+            None,
+            (1, 1),
+            (0, 3),
+            (1, 1),
+            1,
+        )
+        x_182 = l_self_modules_features_modules_12_modules_branch2_modules_4_modules_conv_parameters_weight_ = (None)
+        x_184 = torch.nn.functional.batch_norm(
+            x_183,
+            l_self_modules_features_modules_12_modules_branch2_modules_4_modules_bn_buffers_running_mean_,
+            l_self_modules_features_modules_12_modules_branch2_modules_4_modules_bn_buffers_running_var_,
+            l_self_modules_features_modules_12_modules_branch2_modules_4_modules_bn_parameters_weight_,
+            l_self_modules_features_modules_12_modules_branch2_modules_4_modules_bn_parameters_bias_,
+            False,
+            0.1,
+            0.001,
+        )
+        x_183 = l_self_modules_features_modules_12_modules_branch2_modules_4_modules_bn_buffers_running_mean_ = l_self_modules_features_modules_12_modules_branch2_modules_4_modules_bn_buffers_running_var_ = l_self_modules_features_modules_12_modules_branch2_modules_4_modules_bn_parameters_weight_ = l_self_modules_features_modules_12_modules_branch2_modules_4_modules_bn_parameters_bias_ = (None)
+        x_185 = torch.nn.functional.relu(x_184, inplace=True)
+        x_184 = None
+        input_6 = torch._C._nn.avg_pool2d(out_8, 3, 1, 1, False, False, None)
+        out_8 = None
+        x_186 = torch.conv2d(
+            input_6,
+            l_self_modules_features_modules_12_modules_branch3_modules_1_modules_conv_parameters_weight_,
+            None,
+            (1, 1),
+            (0, 0),
+            (1, 1),
+            1,
+        )
+        input_6 = l_self_modules_features_modules_12_modules_branch3_modules_1_modules_conv_parameters_weight_ = (None)
+        x_187 = torch.nn.functional.batch_norm(
+            x_186,
+            l_self_modules_features_modules_12_modules_branch3_modules_1_modules_bn_buffers_running_mean_,
+            l_self_modules_features_modules_12_modules_branch3_modules_1_modules_bn_buffers_running_var_,
+            l_self_modules_features_modules_12_modules_branch3_modules_1_modules_bn_parameters_weight_,
+            l_self_modules_features_modules_12_modules_branch3_modules_1_modules_bn_parameters_bias_,
+            False,
+            0.1,
+            0.001,
+        )
+        x_186 = l_self_modules_features_modules_12_modules_branch3_modules_1_modules_bn_buffers_running_mean_ = l_self_modules_features_modules_12_modules_branch3_modules_1_modules_bn_buffers_running_var_ = l_self_modules_features_modules_12_modules_branch3_modules_1_modules_bn_parameters_weight_ = l_self_modules_features_modules_12_modules_branch3_modules_1_modules_bn_parameters_bias_ = (None)
+        x_188 = torch.nn.functional.relu(x_187, inplace=True)
+        x_187 = None
+        out_9 = torch.cat((x_161, x_170, x_185, x_188), 1)
+        x_161 = x_170 = x_185 = x_188 = None
+        x_189 = torch.conv2d(
+            out_9,
+            l_self_modules_features_modules_13_modules_branch0_modules_conv_parameters_weight_,
+            None,
+            (1, 1),
+            (0, 0),
+            (1, 1),
+            1,
+        )
+        l_self_modules_features_modules_13_modules_branch0_modules_conv_parameters_weight_ = (
+            None
+        )
+        x_190 = torch.nn.functional.batch_norm(
+            x_189,
+            l_self_modules_features_modules_13_modules_branch0_modules_bn_buffers_running_mean_,
+            l_self_modules_features_modules_13_modules_branch0_modules_bn_buffers_running_var_,
+            l_self_modules_features_modules_13_modules_branch0_modules_bn_parameters_weight_,
+            l_self_modules_features_modules_13_modules_branch0_modules_bn_parameters_bias_,
+            False,
+            0.1,
+            0.001,
+        )
+        x_189 = l_self_modules_features_modules_13_modules_branch0_modules_bn_buffers_running_mean_ = l_self_modules_features_modules_13_modules_branch0_modules_bn_buffers_running_var_ = l_self_modules_features_modules_13_modules_branch0_modules_bn_parameters_weight_ = l_self_modules_features_modules_13_modules_branch0_modules_bn_parameters_bias_ = (None)
+        x_191 = torch.nn.functional.relu(x_190, inplace=True)
+        x_190 = None
+        x_192 = torch.conv2d(
+            out_9,
+            l_self_modules_features_modules_13_modules_branch1_modules_0_modules_conv_parameters_weight_,
+            None,
+            (1, 1),
+            (0, 0),
+            (1, 1),
+            1,
+        )
+        l_self_modules_features_modules_13_modules_branch1_modules_0_modules_conv_parameters_weight_ = (
+            None
+        )
+        x_193 = torch.nn.functional.batch_norm(
+            x_192,
+            l_self_modules_features_modules_13_modules_branch1_modules_0_modules_bn_buffers_running_mean_,
+            l_self_modules_features_modules_13_modules_branch1_modules_0_modules_bn_buffers_running_var_,
+            l_self_modules_features_modules_13_modules_branch1_modules_0_modules_bn_parameters_weight_,
+            l_self_modules_features_modules_13_modules_branch1_modules_0_modules_bn_parameters_bias_,
+            False,
+            0.1,
+            0.001,
+        )
+        x_192 = l_self_modules_features_modules_13_modules_branch1_modules_0_modules_bn_buffers_running_mean_ = l_self_modules_features_modules_13_modules_branch1_modules_0_modules_bn_buffers_running_var_ = l_self_modules_features_modules_13_modules_branch1_modules_0_modules_bn_parameters_weight_ = l_self_modules_features_modules_13_modules_branch1_modules_0_modules_bn_parameters_bias_ = (None)
+        x_194 = torch.nn.functional.relu(x_193, inplace=True)
+        x_193 = None
+        x_195 = torch.conv2d(
+            x_194,
+            l_self_modules_features_modules_13_modules_branch1_modules_1_modules_conv_parameters_weight_,
+            None,
+            (1, 1),
+            (0, 3),
+            (1, 1),
+            1,
+        )
+        x_194 = l_self_modules_features_modules_13_modules_branch1_modules_1_modules_conv_parameters_weight_ = (None)
+        x_196 = torch.nn.functional.batch_norm(
+            x_195,
+            l_self_modules_features_modules_13_modules_branch1_modules_1_modules_bn_buffers_running_mean_,
+            l_self_modules_features_modules_13_modules_branch1_modules_1_modules_bn_buffers_running_var_,
+            l_self_modules_features_modules_13_modules_branch1_modules_1_modules_bn_parameters_weight_,
+            l_self_modules_features_modules_13_modules_branch1_modules_1_modules_bn_parameters_bias_,
+            False,
+            0.1,
+            0.001,
+        )
+        x_195 = l_self_modules_features_modules_13_modules_branch1_modules_1_modules_bn_buffers_running_mean_ = l_self_modules_features_modules_13_modules_branch1_modules_1_modules_bn_buffers_running_var_ = l_self_modules_features_modules_13_modules_branch1_modules_1_modules_bn_parameters_weight_ = l_self_modules_features_modules_13_modules_branch1_modules_1_modules_bn_parameters_bias_ = (None)
+        x_197 = torch.nn.functional.relu(x_196, inplace=True)
+        x_196 = None
+        x_198 = torch.conv2d(
+            x_197,
+            l_self_modules_features_modules_13_modules_branch1_modules_2_modules_conv_parameters_weight_,
+            None,
+            (1, 1),
+            (3, 0),
+            (1, 1),
+            1,
+        )
+        x_197 = l_self_modules_features_modules_13_modules_branch1_modules_2_modules_conv_parameters_weight_ = (None)
+        x_199 = torch.nn.functional.batch_norm(
+            x_198,
+            l_self_modules_features_modules_13_modules_branch1_modules_2_modules_bn_buffers_running_mean_,
+            l_self_modules_features_modules_13_modules_branch1_modules_2_modules_bn_buffers_running_var_,
+            l_self_modules_features_modules_13_modules_branch1_modules_2_modules_bn_parameters_weight_,
+            l_self_modules_features_modules_13_modules_branch1_modules_2_modules_bn_parameters_bias_,
+            False,
+            0.1,
+            0.001,
+        )
+        x_198 = l_self_modules_features_modules_13_modules_branch1_modules_2_modules_bn_buffers_running_mean_ = l_self_modules_features_modules_13_modules_branch1_modules_2_modules_bn_buffers_running_var_ = l_self_modules_features_modules_13_modules_branch1_modules_2_modules_bn_parameters_weight_ = l_self_modules_features_modules_13_modules_branch1_modules_2_modules_bn_parameters_bias_ = (None)
+        x_200 = torch.nn.functional.relu(x_199, inplace=True)
+        x_199 = None
+        x_201 = torch.conv2d(
+            out_9,
+            l_self_modules_features_modules_13_modules_branch2_modules_0_modules_conv_parameters_weight_,
+            None,
+            (1, 1),
+            (0, 0),
+            (1, 1),
+            1,
+        )
+        l_self_modules_features_modules_13_modules_branch2_modules_0_modules_conv_parameters_weight_ = (
+            None
+        )
+        x_202 = torch.nn.functional.batch_norm(
+            x_201,
+            l_self_modules_features_modules_13_modules_branch2_modules_0_modules_bn_buffers_running_mean_,
+            l_self_modules_features_modules_13_modules_branch2_modules_0_modules_bn_buffers_running_var_,
+            l_self_modules_features_modules_13_modules_branch2_modules_0_modules_bn_parameters_weight_,
+            l_self_modules_features_modules_13_modules_branch2_modules_0_modules_bn_parameters_bias_,
+            False,
+            0.1,
+            0.001,
+        )
+        x_201 = l_self_modules_features_modules_13_modules_branch2_modules_0_modules_bn_buffers_running_mean_ = l_self_modules_features_modules_13_modules_branch2_modules_0_modules_bn_buffers_running_var_ = l_self_modules_features_modules_13_modules_branch2_modules_0_modules_bn_parameters_weight_ = l_self_modules_features_modules_13_modules_branch2_modules_0_modules_bn_parameters_bias_ = (None)
+        x_203 = torch.nn.functional.relu(x_202, inplace=True)
+        x_202 = None
+        x_204 = torch.conv2d(
+            x_203,
+            l_self_modules_features_modules_13_modules_branch2_modules_1_modules_conv_parameters_weight_,
+            None,
+            (1, 1),
+            (3, 0),
+            (1, 1),
+            1,
+        )
+        x_203 = l_self_modules_features_modules_13_modules_branch2_modules_1_modules_conv_parameters_weight_ = (None)
+        x_205 = torch.nn.functional.batch_norm(
+            x_204,
+            l_self_modules_features_modules_13_modules_branch2_modules_1_modules_bn_buffers_running_mean_,
+            l_self_modules_features_modules_13_modules_branch2_modules_1_modules_bn_buffers_running_var_,
+            l_self_modules_features_modules_13_modules_branch2_modules_1_modules_bn_parameters_weight_,
+            l_self_modules_features_modules_13_modules_branch2_modules_1_modules_bn_parameters_bias_,
+            False,
+            0.1,
+            0.001,
+        )
+        x_204 = l_self_modules_features_modules_13_modules_branch2_modules_1_modules_bn_buffers_running_mean_ = l_self_modules_features_modules_13_modules_branch2_modules_1_modules_bn_buffers_running_var_ = l_self_modules_features_modules_13_modules_branch2_modules_1_modules_bn_parameters_weight_ = l_self_modules_features_modules_13_modules_branch2_modules_1_modules_bn_parameters_bias_ = (None)
+        x_206 = torch.nn.functional.relu(x_205, inplace=True)
+        x_205 = None
+        x_207 = torch.conv2d(
+            x_206,
+            l_self_modules_features_modules_13_modules_branch2_modules_2_modules_conv_parameters_weight_,
+            None,
+            (1, 1),
+            (0, 3),
+            (1, 1),
+            1,
+        )
+        x_206 = l_self_modules_features_modules_13_modules_branch2_modules_2_modules_conv_parameters_weight_ = (None)
+        x_208 = torch.nn.functional.batch_norm(
+            x_207,
+            l_self_modules_features_modules_13_modules_branch2_modules_2_modules_bn_buffers_running_mean_,
+            l_self_modules_features_modules_13_modules_branch2_modules_2_modules_bn_buffers_running_var_,
+            l_self_modules_features_modules_13_modules_branch2_modules_2_modules_bn_parameters_weight_,
+            l_self_modules_features_modules_13_modules_branch2_modules_2_modules_bn_parameters_bias_,
+            False,
+            0.1,
+            0.001,
+        )
+        x_207 = l_self_modules_features_modules_13_modules_branch2_modules_2_modules_bn_buffers_running_mean_ = l_self_modules_features_modules_13_modules_branch2_modules_2_modules_bn_buffers_running_var_ = l_self_modules_features_modules_13_modules_branch2_modules_2_modules_bn_parameters_weight_ = l_self_modules_features_modules_13_modules_branch2_modules_2_modules_bn_parameters_bias_ = (None)
+        x_209 = torch.nn.functional.relu(x_208, inplace=True)
+        x_208 = None
+        x_210 = torch.conv2d(
+            x_209,
+            l_self_modules_features_modules_13_modules_branch2_modules_3_modules_conv_parameters_weight_,
+            None,
+            (1, 1),
+            (3, 0),
+            (1, 1),
+            1,
+        )
+        x_209 = l_self_modules_features_modules_13_modules_branch2_modules_3_modules_conv_parameters_weight_ = (None)
+        x_211 = torch.nn.functional.batch_norm(
+            x_210,
+            l_self_modules_features_modules_13_modules_branch2_modules_3_modules_bn_buffers_running_mean_,
+            l_self_modules_features_modules_13_modules_branch2_modules_3_modules_bn_buffers_running_var_,
+            l_self_modules_features_modules_13_modules_branch2_modules_3_modules_bn_parameters_weight_,
+            l_self_modules_features_modules_13_modules_branch2_modules_3_modules_bn_parameters_bias_,
+            False,
+            0.1,
+            0.001,
+        )
+        x_210 = l_self_modules_features_modules_13_modules_branch2_modules_3_modules_bn_buffers_running_mean_ = l_self_modules_features_modules_13_modules_branch2_modules_3_modules_bn_buffers_running_var_ = l_self_modules_features_modules_13_modules_branch2_modules_3_modules_bn_parameters_weight_ = l_self_modules_features_modules_13_modules_branch2_modules_3_modules_bn_parameters_bias_ = (None)
+        x_212 = torch.nn.functional.relu(x_211, inplace=True)
+        x_211 = None
+        x_213 = torch.conv2d(
+            x_212,
+            l_self_modules_features_modules_13_modules_branch2_modules_4_modules_conv_parameters_weight_,
+            None,
+            (1, 1),
+            (0, 3),
+            (1, 1),
+            1,
+        )
+        x_212 = l_self_modules_features_modules_13_modules_branch2_modules_4_modules_conv_parameters_weight_ = (None)
+        x_214 = torch.nn.functional.batch_norm(
+            x_213,
+            l_self_modules_features_modules_13_modules_branch2_modules_4_modules_bn_buffers_running_mean_,
+            l_self_modules_features_modules_13_modules_branch2_modules_4_modules_bn_buffers_running_var_,
+            l_self_modules_features_modules_13_modules_branch2_modules_4_modules_bn_parameters_weight_,
+            l_self_modules_features_modules_13_modules_branch2_modules_4_modules_bn_parameters_bias_,
+            False,
+            0.1,
+            0.001,
+        )
+        x_213 = l_self_modules_features_modules_13_modules_branch2_modules_4_modules_bn_buffers_running_mean_ = l_self_modules_features_modules_13_modules_branch2_modules_4_modules_bn_buffers_running_var_ = l_self_modules_features_modules_13_modules_branch2_modules_4_modules_bn_parameters_weight_ = l_self_modules_features_modules_13_modules_branch2_modules_4_modules_bn_parameters_bias_ = (None)
+        x_215 = torch.nn.functional.relu(x_214, inplace=True)
+        x_214 = None
+        input_7 = torch._C._nn.avg_pool2d(out_9, 3, 1, 1, False, False, None)
+        out_9 = None
+        x_216 = torch.conv2d(
+            input_7,
+            l_self_modules_features_modules_13_modules_branch3_modules_1_modules_conv_parameters_weight_,
+            None,
+            (1, 1),
+            (0, 0),
+            (1, 1),
+            1,
+        )
+        input_7 = l_self_modules_features_modules_13_modules_branch3_modules_1_modules_conv_parameters_weight_ = (None)
+        x_217 = torch.nn.functional.batch_norm(
+            x_216,
+            l_self_modules_features_modules_13_modules_branch3_modules_1_modules_bn_buffers_running_mean_,
+            l_self_modules_features_modules_13_modules_branch3_modules_1_modules_bn_buffers_running_var_,
+            l_self_modules_features_modules_13_modules_branch3_modules_1_modules_bn_parameters_weight_,
+            l_self_modules_features_modules_13_modules_branch3_modules_1_modules_bn_parameters_bias_,
+            False,
+            0.1,
+            0.001,
+        )
+        x_216 = l_self_modules_features_modules_13_modules_branch3_modules_1_modules_bn_buffers_running_mean_ = l_self_modules_features_modules_13_modules_branch3_modules_1_modules_bn_buffers_running_var_ = l_self_modules_features_modules_13_modules_branch3_modules_1_modules_bn_parameters_weight_ = l_self_modules_features_modules_13_modules_branch3_modules_1_modules_bn_parameters_bias_ = (None)
+        x_218 = torch.nn.functional.relu(x_217, inplace=True)
+        x_217 = None
+        out_10 = torch.cat((x_191, x_200, x_215, x_218), 1)
+        x_191 = x_200 = x_215 = x_218 = None
+        x_219 = torch.conv2d(
+            out_10,
+            l_self_modules_features_modules_14_modules_branch0_modules_conv_parameters_weight_,
+            None,
+            (1, 1),
+            (0, 0),
+            (1, 1),
+            1,
+        )
+        l_self_modules_features_modules_14_modules_branch0_modules_conv_parameters_weight_ = (
+            None
+        )
+        x_220 = torch.nn.functional.batch_norm(
+            x_219,
+            l_self_modules_features_modules_14_modules_branch0_modules_bn_buffers_running_mean_,
+            l_self_modules_features_modules_14_modules_branch0_modules_bn_buffers_running_var_,
+            l_self_modules_features_modules_14_modules_branch0_modules_bn_parameters_weight_,
+            l_self_modules_features_modules_14_modules_branch0_modules_bn_parameters_bias_,
+            False,
+            0.1,
+            0.001,
+        )
+        x_219 = l_self_modules_features_modules_14_modules_branch0_modules_bn_buffers_running_mean_ = l_self_modules_features_modules_14_modules_branch0_modules_bn_buffers_running_var_ = l_self_modules_features_modules_14_modules_branch0_modules_bn_parameters_weight_ = l_self_modules_features_modules_14_modules_branch0_modules_bn_parameters_bias_ = (None)
+        x_221 = torch.nn.functional.relu(x_220, inplace=True)
+        x_220 = None
+        x_222 = torch.conv2d(
+            out_10,
+            l_self_modules_features_modules_14_modules_branch1_modules_0_modules_conv_parameters_weight_,
+            None,
+            (1, 1),
+            (0, 0),
+            (1, 1),
+            1,
+        )
+        l_self_modules_features_modules_14_modules_branch1_modules_0_modules_conv_parameters_weight_ = (
+            None
+        )
+        x_223 = torch.nn.functional.batch_norm(
+            x_222,
+            l_self_modules_features_modules_14_modules_branch1_modules_0_modules_bn_buffers_running_mean_,
+            l_self_modules_features_modules_14_modules_branch1_modules_0_modules_bn_buffers_running_var_,
+            l_self_modules_features_modules_14_modules_branch1_modules_0_modules_bn_parameters_weight_,
+            l_self_modules_features_modules_14_modules_branch1_modules_0_modules_bn_parameters_bias_,
+            False,
+            0.1,
+            0.001,
+        )
+        x_222 = l_self_modules_features_modules_14_modules_branch1_modules_0_modules_bn_buffers_running_mean_ = l_self_modules_features_modules_14_modules_branch1_modules_0_modules_bn_buffers_running_var_ = l_self_modules_features_modules_14_modules_branch1_modules_0_modules_bn_parameters_weight_ = l_self_modules_features_modules_14_modules_branch1_modules_0_modules_bn_parameters_bias_ = (None)
+        x_224 = torch.nn.functional.relu(x_223, inplace=True)
+        x_223 = None
+        x_225 = torch.conv2d(
+            x_224,
+            l_self_modules_features_modules_14_modules_branch1_modules_1_modules_conv_parameters_weight_,
+            None,
+            (1, 1),
+            (0, 3),
+            (1, 1),
+            1,
+        )
+        x_224 = l_self_modules_features_modules_14_modules_branch1_modules_1_modules_conv_parameters_weight_ = (None)
+        x_226 = torch.nn.functional.batch_norm(
+            x_225,
+            l_self_modules_features_modules_14_modules_branch1_modules_1_modules_bn_buffers_running_mean_,
+            l_self_modules_features_modules_14_modules_branch1_modules_1_modules_bn_buffers_running_var_,
+            l_self_modules_features_modules_14_modules_branch1_modules_1_modules_bn_parameters_weight_,
+            l_self_modules_features_modules_14_modules_branch1_modules_1_modules_bn_parameters_bias_,
+            False,
+            0.1,
+            0.001,
+        )
+        x_225 = l_self_modules_features_modules_14_modules_branch1_modules_1_modules_bn_buffers_running_mean_ = l_self_modules_features_modules_14_modules_branch1_modules_1_modules_bn_buffers_running_var_ = l_self_modules_features_modules_14_modules_branch1_modules_1_modules_bn_parameters_weight_ = l_self_modules_features_modules_14_modules_branch1_modules_1_modules_bn_parameters_bias_ = (None)
+        x_227 = torch.nn.functional.relu(x_226, inplace=True)
+        x_226 = None
+        x_228 = torch.conv2d(
+            x_227,
+            l_self_modules_features_modules_14_modules_branch1_modules_2_modules_conv_parameters_weight_,
+            None,
+            (1, 1),
+            (3, 0),
+            (1, 1),
+            1,
+        )
+        x_227 = l_self_modules_features_modules_14_modules_branch1_modules_2_modules_conv_parameters_weight_ = (None)
+        x_229 = torch.nn.functional.batch_norm(
+            x_228,
+            l_self_modules_features_modules_14_modules_branch1_modules_2_modules_bn_buffers_running_mean_,
+            l_self_modules_features_modules_14_modules_branch1_modules_2_modules_bn_buffers_running_var_,
+            l_self_modules_features_modules_14_modules_branch1_modules_2_modules_bn_parameters_weight_,
+            l_self_modules_features_modules_14_modules_branch1_modules_2_modules_bn_parameters_bias_,
+            False,
+            0.1,
+            0.001,
+        )
+        x_228 = l_self_modules_features_modules_14_modules_branch1_modules_2_modules_bn_buffers_running_mean_ = l_self_modules_features_modules_14_modules_branch1_modules_2_modules_bn_buffers_running_var_ = l_self_modules_features_modules_14_modules_branch1_modules_2_modules_bn_parameters_weight_ = l_self_modules_features_modules_14_modules_branch1_modules_2_modules_bn_parameters_bias_ = (None)
+        x_230 = torch.nn.functional.relu(x_229, inplace=True)
+        x_229 = None
+        x_231 = torch.conv2d(
+            out_10,
+            l_self_modules_features_modules_14_modules_branch2_modules_0_modules_conv_parameters_weight_,
+            None,
+            (1, 1),
+            (0, 0),
+            (1, 1),
+            1,
+        )
+        l_self_modules_features_modules_14_modules_branch2_modules_0_modules_conv_parameters_weight_ = (
+            None
+        )
+        x_232 = torch.nn.functional.batch_norm(
+            x_231,
+            l_self_modules_features_modules_14_modules_branch2_modules_0_modules_bn_buffers_running_mean_,
+            l_self_modules_features_modules_14_modules_branch2_modules_0_modules_bn_buffers_running_var_,
+            l_self_modules_features_modules_14_modules_branch2_modules_0_modules_bn_parameters_weight_,
+            l_self_modules_features_modules_14_modules_branch2_modules_0_modules_bn_parameters_bias_,
+            False,
+            0.1,
+            0.001,
+        )
+        x_231 = l_self_modules_features_modules_14_modules_branch2_modules_0_modules_bn_buffers_running_mean_ = l_self_modules_features_modules_14_modules_branch2_modules_0_modules_bn_buffers_running_var_ = l_self_modules_features_modules_14_modules_branch2_modules_0_modules_bn_parameters_weight_ = l_self_modules_features_modules_14_modules_branch2_modules_0_modules_bn_parameters_bias_ = (None)
+        x_233 = torch.nn.functional.relu(x_232, inplace=True)
+        x_232 = None
+        x_234 = torch.conv2d(
+            x_233,
+            l_self_modules_features_modules_14_modules_branch2_modules_1_modules_conv_parameters_weight_,
+            None,
+            (1, 1),
+            (3, 0),
+            (1, 1),
+            1,
+        )
+        x_233 = l_self_modules_features_modules_14_modules_branch2_modules_1_modules_conv_parameters_weight_ = (None)
+        x_235 = torch.nn.functional.batch_norm(
+            x_234,
+            l_self_modules_features_modules_14_modules_branch2_modules_1_modules_bn_buffers_running_mean_,
+            l_self_modules_features_modules_14_modules_branch2_modules_1_modules_bn_buffers_running_var_,
+            l_self_modules_features_modules_14_modules_branch2_modules_1_modules_bn_parameters_weight_,
+            l_self_modules_features_modules_14_modules_branch2_modules_1_modules_bn_parameters_bias_,
+            False,
+            0.1,
+            0.001,
+        )
+        x_234 = l_self_modules_features_modules_14_modules_branch2_modules_1_modules_bn_buffers_running_mean_ = l_self_modules_features_modules_14_modules_branch2_modules_1_modules_bn_buffers_running_var_ = l_self_modules_features_modules_14_modules_branch2_modules_1_modules_bn_parameters_weight_ = l_self_modules_features_modules_14_modules_branch2_modules_1_modules_bn_parameters_bias_ = (None)
+        x_236 = torch.nn.functional.relu(x_235, inplace=True)
+        x_235 = None
+        x_237 = torch.conv2d(
+            x_236,
+            l_self_modules_features_modules_14_modules_branch2_modules_2_modules_conv_parameters_weight_,
+            None,
+            (1, 1),
+            (0, 3),
+            (1, 1),
+            1,
+        )
+        x_236 = l_self_modules_features_modules_14_modules_branch2_modules_2_modules_conv_parameters_weight_ = (None)
+        x_238 = torch.nn.functional.batch_norm(
+            x_237,
+            l_self_modules_features_modules_14_modules_branch2_modules_2_modules_bn_buffers_running_mean_,
+            l_self_modules_features_modules_14_modules_branch2_modules_2_modules_bn_buffers_running_var_,
+            l_self_modules_features_modules_14_modules_branch2_modules_2_modules_bn_parameters_weight_,
+            l_self_modules_features_modules_14_modules_branch2_modules_2_modules_bn_parameters_bias_,
+            False,
+            0.1,
+            0.001,
+        )
+        x_237 = l_self_modules_features_modules_14_modules_branch2_modules_2_modules_bn_buffers_running_mean_ = l_self_modules_features_modules_14_modules_branch2_modules_2_modules_bn_buffers_running_var_ = l_self_modules_features_modules_14_modules_branch2_modules_2_modules_bn_parameters_weight_ = l_self_modules_features_modules_14_modules_branch2_modules_2_modules_bn_parameters_bias_ = (None)
+        x_239 = torch.nn.functional.relu(x_238, inplace=True)
+        x_238 = None
+        x_240 = torch.conv2d(
+            x_239,
+            l_self_modules_features_modules_14_modules_branch2_modules_3_modules_conv_parameters_weight_,
+            None,
+            (1, 1),
+            (3, 0),
+            (1, 1),
+            1,
+        )
+        x_239 = l_self_modules_features_modules_14_modules_branch2_modules_3_modules_conv_parameters_weight_ = (None)
+        x_241 = torch.nn.functional.batch_norm(
+            x_240,
+            l_self_modules_features_modules_14_modules_branch2_modules_3_modules_bn_buffers_running_mean_,
+            l_self_modules_features_modules_14_modules_branch2_modules_3_modules_bn_buffers_running_var_,
+            l_self_modules_features_modules_14_modules_branch2_modules_3_modules_bn_parameters_weight_,
+            l_self_modules_features_modules_14_modules_branch2_modules_3_modules_bn_parameters_bias_,
+            False,
+            0.1,
+            0.001,
+        )
+        x_240 = l_self_modules_features_modules_14_modules_branch2_modules_3_modules_bn_buffers_running_mean_ = l_self_modules_features_modules_14_modules_branch2_modules_3_modules_bn_buffers_running_var_ = l_self_modules_features_modules_14_modules_branch2_modules_3_modules_bn_parameters_weight_ = l_self_modules_features_modules_14_modules_branch2_modules_3_modules_bn_parameters_bias_ = (None)
+        x_242 = torch.nn.functional.relu(x_241, inplace=True)
+        x_241 = None
+        x_243 = torch.conv2d(
+            x_242,
+            l_self_modules_features_modules_14_modules_branch2_modules_4_modules_conv_parameters_weight_,
+            None,
+            (1, 1),
+            (0, 3),
+            (1, 1),
+            1,
+        )
+        x_242 = l_self_modules_features_modules_14_modules_branch2_modules_4_modules_conv_parameters_weight_ = (None)
+        x_244 = torch.nn.functional.batch_norm(
+            x_243,
+            l_self_modules_features_modules_14_modules_branch2_modules_4_modules_bn_buffers_running_mean_,
+            l_self_modules_features_modules_14_modules_branch2_modules_4_modules_bn_buffers_running_var_,
+            l_self_modules_features_modules_14_modules_branch2_modules_4_modules_bn_parameters_weight_,
+            l_self_modules_features_modules_14_modules_branch2_modules_4_modules_bn_parameters_bias_,
+            False,
+            0.1,
+            0.001,
+        )
+        x_243 = l_self_modules_features_modules_14_modules_branch2_modules_4_modules_bn_buffers_running_mean_ = l_self_modules_features_modules_14_modules_branch2_modules_4_modules_bn_buffers_running_var_ = l_self_modules_features_modules_14_modules_branch2_modules_4_modules_bn_parameters_weight_ = l_self_modules_features_modules_14_modules_branch2_modules_4_modules_bn_parameters_bias_ = (None)
+        x_245 = torch.nn.functional.relu(x_244, inplace=True)
+        x_244 = None
+        input_8 = torch._C._nn.avg_pool2d(out_10, 3, 1, 1, False, False, None)
+        out_10 = None
+        x_246 = torch.conv2d(
+            input_8,
+            l_self_modules_features_modules_14_modules_branch3_modules_1_modules_conv_parameters_weight_,
+            None,
+            (1, 1),
+            (0, 0),
+            (1, 1),
+            1,
+        )
+        input_8 = l_self_modules_features_modules_14_modules_branch3_modules_1_modules_conv_parameters_weight_ = (None)
+        x_247 = torch.nn.functional.batch_norm(
+            x_246,
+            l_self_modules_features_modules_14_modules_branch3_modules_1_modules_bn_buffers_running_mean_,
+            l_self_modules_features_modules_14_modules_branch3_modules_1_modules_bn_buffers_running_var_,
+            l_self_modules_features_modules_14_modules_branch3_modules_1_modules_bn_parameters_weight_,
+            l_self_modules_features_modules_14_modules_branch3_modules_1_modules_bn_parameters_bias_,
+            False,
+            0.1,
+            0.001,
+        )
+        x_246 = l_self_modules_features_modules_14_modules_branch3_modules_1_modules_bn_buffers_running_mean_ = l_self_modules_features_modules_14_modules_branch3_modules_1_modules_bn_buffers_running_var_ = l_self_modules_features_modules_14_modules_branch3_modules_1_modules_bn_parameters_weight_ = l_self_modules_features_modules_14_modules_branch3_modules_1_modules_bn_parameters_bias_ = (None)
+        x_248 = torch.nn.functional.relu(x_247, inplace=True)
+        x_247 = None
+        out_11 = torch.cat((x_221, x_230, x_245, x_248), 1)
+        x_221 = x_230 = x_245 = x_248 = None
+        x_249 = torch.conv2d(
+            out_11,
+            l_self_modules_features_modules_15_modules_branch0_modules_conv_parameters_weight_,
+            None,
+            (1, 1),
+            (0, 0),
+            (1, 1),
+            1,
+        )
+        l_self_modules_features_modules_15_modules_branch0_modules_conv_parameters_weight_ = (
+            None
+        )
+        x_250 = torch.nn.functional.batch_norm(
+            x_249,
+            l_self_modules_features_modules_15_modules_branch0_modules_bn_buffers_running_mean_,
+            l_self_modules_features_modules_15_modules_branch0_modules_bn_buffers_running_var_,
+            l_self_modules_features_modules_15_modules_branch0_modules_bn_parameters_weight_,
+            l_self_modules_features_modules_15_modules_branch0_modules_bn_parameters_bias_,
+            False,
+            0.1,
+            0.001,
+        )
+        x_249 = l_self_modules_features_modules_15_modules_branch0_modules_bn_buffers_running_mean_ = l_self_modules_features_modules_15_modules_branch0_modules_bn_buffers_running_var_ = l_self_modules_features_modules_15_modules_branch0_modules_bn_parameters_weight_ = l_self_modules_features_modules_15_modules_branch0_modules_bn_parameters_bias_ = (None)
+        x_251 = torch.nn.functional.relu(x_250, inplace=True)
+        x_250 = None
+        x_252 = torch.conv2d(
+            out_11,
+            l_self_modules_features_modules_15_modules_branch1_modules_0_modules_conv_parameters_weight_,
+            None,
+            (1, 1),
+            (0, 0),
+            (1, 1),
+            1,
+        )
+        l_self_modules_features_modules_15_modules_branch1_modules_0_modules_conv_parameters_weight_ = (
+            None
+        )
+        x_253 = torch.nn.functional.batch_norm(
+            x_252,
+            l_self_modules_features_modules_15_modules_branch1_modules_0_modules_bn_buffers_running_mean_,
+            l_self_modules_features_modules_15_modules_branch1_modules_0_modules_bn_buffers_running_var_,
+            l_self_modules_features_modules_15_modules_branch1_modules_0_modules_bn_parameters_weight_,
+            l_self_modules_features_modules_15_modules_branch1_modules_0_modules_bn_parameters_bias_,
+            False,
+            0.1,
+            0.001,
+        )
+        x_252 = l_self_modules_features_modules_15_modules_branch1_modules_0_modules_bn_buffers_running_mean_ = l_self_modules_features_modules_15_modules_branch1_modules_0_modules_bn_buffers_running_var_ = l_self_modules_features_modules_15_modules_branch1_modules_0_modules_bn_parameters_weight_ = l_self_modules_features_modules_15_modules_branch1_modules_0_modules_bn_parameters_bias_ = (None)
+        x_254 = torch.nn.functional.relu(x_253, inplace=True)
+        x_253 = None
+        x_255 = torch.conv2d(
+            x_254,
+            l_self_modules_features_modules_15_modules_branch1_modules_1_modules_conv_parameters_weight_,
+            None,
+            (1, 1),
+            (0, 3),
+            (1, 1),
+            1,
+        )
+        x_254 = l_self_modules_features_modules_15_modules_branch1_modules_1_modules_conv_parameters_weight_ = (None)
+        x_256 = torch.nn.functional.batch_norm(
+            x_255,
+            l_self_modules_features_modules_15_modules_branch1_modules_1_modules_bn_buffers_running_mean_,
+            l_self_modules_features_modules_15_modules_branch1_modules_1_modules_bn_buffers_running_var_,
+            l_self_modules_features_modules_15_modules_branch1_modules_1_modules_bn_parameters_weight_,
+            l_self_modules_features_modules_15_modules_branch1_modules_1_modules_bn_parameters_bias_,
+            False,
+            0.1,
+            0.001,
+        )
+        x_255 = l_self_modules_features_modules_15_modules_branch1_modules_1_modules_bn_buffers_running_mean_ = l_self_modules_features_modules_15_modules_branch1_modules_1_modules_bn_buffers_running_var_ = l_self_modules_features_modules_15_modules_branch1_modules_1_modules_bn_parameters_weight_ = l_self_modules_features_modules_15_modules_branch1_modules_1_modules_bn_parameters_bias_ = (None)
+        x_257 = torch.nn.functional.relu(x_256, inplace=True)
+        x_256 = None
+        x_258 = torch.conv2d(
+            x_257,
+            l_self_modules_features_modules_15_modules_branch1_modules_2_modules_conv_parameters_weight_,
+            None,
+            (1, 1),
+            (3, 0),
+            (1, 1),
+            1,
+        )
+        x_257 = l_self_modules_features_modules_15_modules_branch1_modules_2_modules_conv_parameters_weight_ = (None)
+        x_259 = torch.nn.functional.batch_norm(
+            x_258,
+            l_self_modules_features_modules_15_modules_branch1_modules_2_modules_bn_buffers_running_mean_,
+            l_self_modules_features_modules_15_modules_branch1_modules_2_modules_bn_buffers_running_var_,
+            l_self_modules_features_modules_15_modules_branch1_modules_2_modules_bn_parameters_weight_,
+            l_self_modules_features_modules_15_modules_branch1_modules_2_modules_bn_parameters_bias_,
+            False,
+            0.1,
+            0.001,
+        )
+        x_258 = l_self_modules_features_modules_15_modules_branch1_modules_2_modules_bn_buffers_running_mean_ = l_self_modules_features_modules_15_modules_branch1_modules_2_modules_bn_buffers_running_var_ = l_self_modules_features_modules_15_modules_branch1_modules_2_modules_bn_parameters_weight_ = l_self_modules_features_modules_15_modules_branch1_modules_2_modules_bn_parameters_bias_ = (None)
+        x_260 = torch.nn.functional.relu(x_259, inplace=True)
+        x_259 = None
+        x_261 = torch.conv2d(
+            out_11,
+            l_self_modules_features_modules_15_modules_branch2_modules_0_modules_conv_parameters_weight_,
+            None,
+            (1, 1),
+            (0, 0),
+            (1, 1),
+            1,
+        )
+        l_self_modules_features_modules_15_modules_branch2_modules_0_modules_conv_parameters_weight_ = (
+            None
+        )
+        x_262 = torch.nn.functional.batch_norm(
+            x_261,
+            l_self_modules_features_modules_15_modules_branch2_modules_0_modules_bn_buffers_running_mean_,
+            l_self_modules_features_modules_15_modules_branch2_modules_0_modules_bn_buffers_running_var_,
+            l_self_modules_features_modules_15_modules_branch2_modules_0_modules_bn_parameters_weight_,
+            l_self_modules_features_modules_15_modules_branch2_modules_0_modules_bn_parameters_bias_,
+            False,
+            0.1,
+            0.001,
+        )
+        x_261 = l_self_modules_features_modules_15_modules_branch2_modules_0_modules_bn_buffers_running_mean_ = l_self_modules_features_modules_15_modules_branch2_modules_0_modules_bn_buffers_running_var_ = l_self_modules_features_modules_15_modules_branch2_modules_0_modules_bn_parameters_weight_ = l_self_modules_features_modules_15_modules_branch2_modules_0_modules_bn_parameters_bias_ = (None)
+        x_263 = torch.nn.functional.relu(x_262, inplace=True)
+        x_262 = None
+        x_264 = torch.conv2d(
+            x_263,
+            l_self_modules_features_modules_15_modules_branch2_modules_1_modules_conv_parameters_weight_,
+            None,
+            (1, 1),
+            (3, 0),
+            (1, 1),
+            1,
+        )
+        x_263 = l_self_modules_features_modules_15_modules_branch2_modules_1_modules_conv_parameters_weight_ = (None)
+        x_265 = torch.nn.functional.batch_norm(
+            x_264,
+            l_self_modules_features_modules_15_modules_branch2_modules_1_modules_bn_buffers_running_mean_,
+            l_self_modules_features_modules_15_modules_branch2_modules_1_modules_bn_buffers_running_var_,
+            l_self_modules_features_modules_15_modules_branch2_modules_1_modules_bn_parameters_weight_,
+            l_self_modules_features_modules_15_modules_branch2_modules_1_modules_bn_parameters_bias_,
+            False,
+            0.1,
+            0.001,
+        )
+        x_264 = l_self_modules_features_modules_15_modules_branch2_modules_1_modules_bn_buffers_running_mean_ = l_self_modules_features_modules_15_modules_branch2_modules_1_modules_bn_buffers_running_var_ = l_self_modules_features_modules_15_modules_branch2_modules_1_modules_bn_parameters_weight_ = l_self_modules_features_modules_15_modules_branch2_modules_1_modules_bn_parameters_bias_ = (None)
+        x_266 = torch.nn.functional.relu(x_265, inplace=True)
+        x_265 = None
+        x_267 = torch.conv2d(
+            x_266,
+            l_self_modules_features_modules_15_modules_branch2_modules_2_modules_conv_parameters_weight_,
+            None,
+            (1, 1),
+            (0, 3),
+            (1, 1),
+            1,
+        )
+        x_266 = l_self_modules_features_modules_15_modules_branch2_modules_2_modules_conv_parameters_weight_ = (None)
+        x_268 = torch.nn.functional.batch_norm(
+            x_267,
+            l_self_modules_features_modules_15_modules_branch2_modules_2_modules_bn_buffers_running_mean_,
+            l_self_modules_features_modules_15_modules_branch2_modules_2_modules_bn_buffers_running_var_,
+            l_self_modules_features_modules_15_modules_branch2_modules_2_modules_bn_parameters_weight_,
+            l_self_modules_features_modules_15_modules_branch2_modules_2_modules_bn_parameters_bias_,
+            False,
+            0.1,
+            0.001,
+        )
+        x_267 = l_self_modules_features_modules_15_modules_branch2_modules_2_modules_bn_buffers_running_mean_ = l_self_modules_features_modules_15_modules_branch2_modules_2_modules_bn_buffers_running_var_ = l_self_modules_features_modules_15_modules_branch2_modules_2_modules_bn_parameters_weight_ = l_self_modules_features_modules_15_modules_branch2_modules_2_modules_bn_parameters_bias_ = (None)
+        x_269 = torch.nn.functional.relu(x_268, inplace=True)
+        x_268 = None
+        x_270 = torch.conv2d(
+            x_269,
+            l_self_modules_features_modules_15_modules_branch2_modules_3_modules_conv_parameters_weight_,
+            None,
+            (1, 1),
+            (3, 0),
+            (1, 1),
+            1,
+        )
+        x_269 = l_self_modules_features_modules_15_modules_branch2_modules_3_modules_conv_parameters_weight_ = (None)
+        x_271 = torch.nn.functional.batch_norm(
+            x_270,
+            l_self_modules_features_modules_15_modules_branch2_modules_3_modules_bn_buffers_running_mean_,
+            l_self_modules_features_modules_15_modules_branch2_modules_3_modules_bn_buffers_running_var_,
+            l_self_modules_features_modules_15_modules_branch2_modules_3_modules_bn_parameters_weight_,
+            l_self_modules_features_modules_15_modules_branch2_modules_3_modules_bn_parameters_bias_,
+            False,
+            0.1,
+            0.001,
+        )
+        x_270 = l_self_modules_features_modules_15_modules_branch2_modules_3_modules_bn_buffers_running_mean_ = l_self_modules_features_modules_15_modules_branch2_modules_3_modules_bn_buffers_running_var_ = l_self_modules_features_modules_15_modules_branch2_modules_3_modules_bn_parameters_weight_ = l_self_modules_features_modules_15_modules_branch2_modules_3_modules_bn_parameters_bias_ = (None)
+        x_272 = torch.nn.functional.relu(x_271, inplace=True)
+        x_271 = None
+        x_273 = torch.conv2d(
+            x_272,
+            l_self_modules_features_modules_15_modules_branch2_modules_4_modules_conv_parameters_weight_,
+            None,
+            (1, 1),
+            (0, 3),
+            (1, 1),
+            1,
+        )
+        x_272 = l_self_modules_features_modules_15_modules_branch2_modules_4_modules_conv_parameters_weight_ = (None)
+        x_274 = torch.nn.functional.batch_norm(
+            x_273,
+            l_self_modules_features_modules_15_modules_branch2_modules_4_modules_bn_buffers_running_mean_,
+            l_self_modules_features_modules_15_modules_branch2_modules_4_modules_bn_buffers_running_var_,
+            l_self_modules_features_modules_15_modules_branch2_modules_4_modules_bn_parameters_weight_,
+            l_self_modules_features_modules_15_modules_branch2_modules_4_modules_bn_parameters_bias_,
+            False,
+            0.1,
+            0.001,
+        )
+        x_273 = l_self_modules_features_modules_15_modules_branch2_modules_4_modules_bn_buffers_running_mean_ = l_self_modules_features_modules_15_modules_branch2_modules_4_modules_bn_buffers_running_var_ = l_self_modules_features_modules_15_modules_branch2_modules_4_modules_bn_parameters_weight_ = l_self_modules_features_modules_15_modules_branch2_modules_4_modules_bn_parameters_bias_ = (None)
+        x_275 = torch.nn.functional.relu(x_274, inplace=True)
+        x_274 = None
+        input_9 = torch._C._nn.avg_pool2d(out_11, 3, 1, 1, False, False, None)
+        out_11 = None
+        x_276 = torch.conv2d(
+            input_9,
+            l_self_modules_features_modules_15_modules_branch3_modules_1_modules_conv_parameters_weight_,
+            None,
+            (1, 1),
+            (0, 0),
+            (1, 1),
+            1,
+        )
+        input_9 = l_self_modules_features_modules_15_modules_branch3_modules_1_modules_conv_parameters_weight_ = (None)
+        x_277 = torch.nn.functional.batch_norm(
+            x_276,
+            l_self_modules_features_modules_15_modules_branch3_modules_1_modules_bn_buffers_running_mean_,
+            l_self_modules_features_modules_15_modules_branch3_modules_1_modules_bn_buffers_running_var_,
+            l_self_modules_features_modules_15_modules_branch3_modules_1_modules_bn_parameters_weight_,
+            l_self_modules_features_modules_15_modules_branch3_modules_1_modules_bn_parameters_bias_,
+            False,
+            0.1,
+            0.001,
+        )
+        x_276 = l_self_modules_features_modules_15_modules_branch3_modules_1_modules_bn_buffers_running_mean_ = l_self_modules_features_modules_15_modules_branch3_modules_1_modules_bn_buffers_running_var_ = l_self_modules_features_modules_15_modules_branch3_modules_1_modules_bn_parameters_weight_ = l_self_modules_features_modules_15_modules_branch3_modules_1_modules_bn_parameters_bias_ = (None)
+        x_278 = torch.nn.functional.relu(x_277, inplace=True)
+        x_277 = None
+        out_12 = torch.cat((x_251, x_260, x_275, x_278), 1)
+        x_251 = x_260 = x_275 = x_278 = None
+        x_279 = torch.conv2d(
+            out_12,
+            l_self_modules_features_modules_16_modules_branch0_modules_conv_parameters_weight_,
+            None,
+            (1, 1),
+            (0, 0),
+            (1, 1),
+            1,
+        )
+        l_self_modules_features_modules_16_modules_branch0_modules_conv_parameters_weight_ = (
+            None
+        )
+        x_280 = torch.nn.functional.batch_norm(
+            x_279,
+            l_self_modules_features_modules_16_modules_branch0_modules_bn_buffers_running_mean_,
+            l_self_modules_features_modules_16_modules_branch0_modules_bn_buffers_running_var_,
+            l_self_modules_features_modules_16_modules_branch0_modules_bn_parameters_weight_,
+            l_self_modules_features_modules_16_modules_branch0_modules_bn_parameters_bias_,
+            False,
+            0.1,
+            0.001,
+        )
+        x_279 = l_self_modules_features_modules_16_modules_branch0_modules_bn_buffers_running_mean_ = l_self_modules_features_modules_16_modules_branch0_modules_bn_buffers_running_var_ = l_self_modules_features_modules_16_modules_branch0_modules_bn_parameters_weight_ = l_self_modules_features_modules_16_modules_branch0_modules_bn_parameters_bias_ = (None)
+        x_281 = torch.nn.functional.relu(x_280, inplace=True)
+        x_280 = None
+        x_282 = torch.conv2d(
+            out_12,
+            l_self_modules_features_modules_16_modules_branch1_modules_0_modules_conv_parameters_weight_,
+            None,
+            (1, 1),
+            (0, 0),
+            (1, 1),
+            1,
+        )
+        l_self_modules_features_modules_16_modules_branch1_modules_0_modules_conv_parameters_weight_ = (
+            None
+        )
+        x_283 = torch.nn.functional.batch_norm(
+            x_282,
+            l_self_modules_features_modules_16_modules_branch1_modules_0_modules_bn_buffers_running_mean_,
+            l_self_modules_features_modules_16_modules_branch1_modules_0_modules_bn_buffers_running_var_,
+            l_self_modules_features_modules_16_modules_branch1_modules_0_modules_bn_parameters_weight_,
+            l_self_modules_features_modules_16_modules_branch1_modules_0_modules_bn_parameters_bias_,
+            False,
+            0.1,
+            0.001,
+        )
+        x_282 = l_self_modules_features_modules_16_modules_branch1_modules_0_modules_bn_buffers_running_mean_ = l_self_modules_features_modules_16_modules_branch1_modules_0_modules_bn_buffers_running_var_ = l_self_modules_features_modules_16_modules_branch1_modules_0_modules_bn_parameters_weight_ = l_self_modules_features_modules_16_modules_branch1_modules_0_modules_bn_parameters_bias_ = (None)
+        x_284 = torch.nn.functional.relu(x_283, inplace=True)
+        x_283 = None
+        x_285 = torch.conv2d(
+            x_284,
+            l_self_modules_features_modules_16_modules_branch1_modules_1_modules_conv_parameters_weight_,
+            None,
+            (1, 1),
+            (0, 3),
+            (1, 1),
+            1,
+        )
+        x_284 = l_self_modules_features_modules_16_modules_branch1_modules_1_modules_conv_parameters_weight_ = (None)
+        x_286 = torch.nn.functional.batch_norm(
+            x_285,
+            l_self_modules_features_modules_16_modules_branch1_modules_1_modules_bn_buffers_running_mean_,
+            l_self_modules_features_modules_16_modules_branch1_modules_1_modules_bn_buffers_running_var_,
+            l_self_modules_features_modules_16_modules_branch1_modules_1_modules_bn_parameters_weight_,
+            l_self_modules_features_modules_16_modules_branch1_modules_1_modules_bn_parameters_bias_,
+            False,
+            0.1,
+            0.001,
+        )
+        x_285 = l_self_modules_features_modules_16_modules_branch1_modules_1_modules_bn_buffers_running_mean_ = l_self_modules_features_modules_16_modules_branch1_modules_1_modules_bn_buffers_running_var_ = l_self_modules_features_modules_16_modules_branch1_modules_1_modules_bn_parameters_weight_ = l_self_modules_features_modules_16_modules_branch1_modules_1_modules_bn_parameters_bias_ = (None)
+        x_287 = torch.nn.functional.relu(x_286, inplace=True)
+        x_286 = None
+        x_288 = torch.conv2d(
+            x_287,
+            l_self_modules_features_modules_16_modules_branch1_modules_2_modules_conv_parameters_weight_,
+            None,
+            (1, 1),
+            (3, 0),
+            (1, 1),
+            1,
+        )
+        x_287 = l_self_modules_features_modules_16_modules_branch1_modules_2_modules_conv_parameters_weight_ = (None)
+        x_289 = torch.nn.functional.batch_norm(
+            x_288,
+            l_self_modules_features_modules_16_modules_branch1_modules_2_modules_bn_buffers_running_mean_,
+            l_self_modules_features_modules_16_modules_branch1_modules_2_modules_bn_buffers_running_var_,
+            l_self_modules_features_modules_16_modules_branch1_modules_2_modules_bn_parameters_weight_,
+            l_self_modules_features_modules_16_modules_branch1_modules_2_modules_bn_parameters_bias_,
+            False,
+            0.1,
+            0.001,
+        )
+        x_288 = l_self_modules_features_modules_16_modules_branch1_modules_2_modules_bn_buffers_running_mean_ = l_self_modules_features_modules_16_modules_branch1_modules_2_modules_bn_buffers_running_var_ = l_self_modules_features_modules_16_modules_branch1_modules_2_modules_bn_parameters_weight_ = l_self_modules_features_modules_16_modules_branch1_modules_2_modules_bn_parameters_bias_ = (None)
+        x_290 = torch.nn.functional.relu(x_289, inplace=True)
+        x_289 = None
+        x_291 = torch.conv2d(
+            out_12,
+            l_self_modules_features_modules_16_modules_branch2_modules_0_modules_conv_parameters_weight_,
+            None,
+            (1, 1),
+            (0, 0),
+            (1, 1),
+            1,
+        )
+        l_self_modules_features_modules_16_modules_branch2_modules_0_modules_conv_parameters_weight_ = (
+            None
+        )
+        x_292 = torch.nn.functional.batch_norm(
+            x_291,
+            l_self_modules_features_modules_16_modules_branch2_modules_0_modules_bn_buffers_running_mean_,
+            l_self_modules_features_modules_16_modules_branch2_modules_0_modules_bn_buffers_running_var_,
+            l_self_modules_features_modules_16_modules_branch2_modules_0_modules_bn_parameters_weight_,
+            l_self_modules_features_modules_16_modules_branch2_modules_0_modules_bn_parameters_bias_,
+            False,
+            0.1,
+            0.001,
+        )
+        x_291 = l_self_modules_features_modules_16_modules_branch2_modules_0_modules_bn_buffers_running_mean_ = l_self_modules_features_modules_16_modules_branch2_modules_0_modules_bn_buffers_running_var_ = l_self_modules_features_modules_16_modules_branch2_modules_0_modules_bn_parameters_weight_ = l_self_modules_features_modules_16_modules_branch2_modules_0_modules_bn_parameters_bias_ = (None)
+        x_293 = torch.nn.functional.relu(x_292, inplace=True)
+        x_292 = None
+        x_294 = torch.conv2d(
+            x_293,
+            l_self_modules_features_modules_16_modules_branch2_modules_1_modules_conv_parameters_weight_,
+            None,
+            (1, 1),
+            (3, 0),
+            (1, 1),
+            1,
+        )
+        x_293 = l_self_modules_features_modules_16_modules_branch2_modules_1_modules_conv_parameters_weight_ = (None)
+        x_295 = torch.nn.functional.batch_norm(
+            x_294,
+            l_self_modules_features_modules_16_modules_branch2_modules_1_modules_bn_buffers_running_mean_,
+            l_self_modules_features_modules_16_modules_branch2_modules_1_modules_bn_buffers_running_var_,
+            l_self_modules_features_modules_16_modules_branch2_modules_1_modules_bn_parameters_weight_,
+            l_self_modules_features_modules_16_modules_branch2_modules_1_modules_bn_parameters_bias_,
+            False,
+            0.1,
+            0.001,
+        )
+        x_294 = l_self_modules_features_modules_16_modules_branch2_modules_1_modules_bn_buffers_running_mean_ = l_self_modules_features_modules_16_modules_branch2_modules_1_modules_bn_buffers_running_var_ = l_self_modules_features_modules_16_modules_branch2_modules_1_modules_bn_parameters_weight_ = l_self_modules_features_modules_16_modules_branch2_modules_1_modules_bn_parameters_bias_ = (None)
+        x_296 = torch.nn.functional.relu(x_295, inplace=True)
+        x_295 = None
+        x_297 = torch.conv2d(
+            x_296,
+            l_self_modules_features_modules_16_modules_branch2_modules_2_modules_conv_parameters_weight_,
+            None,
+            (1, 1),
+            (0, 3),
+            (1, 1),
+            1,
+        )
+        x_296 = l_self_modules_features_modules_16_modules_branch2_modules_2_modules_conv_parameters_weight_ = (None)
+        x_298 = torch.nn.functional.batch_norm(
+            x_297,
+            l_self_modules_features_modules_16_modules_branch2_modules_2_modules_bn_buffers_running_mean_,
+            l_self_modules_features_modules_16_modules_branch2_modules_2_modules_bn_buffers_running_var_,
+            l_self_modules_features_modules_16_modules_branch2_modules_2_modules_bn_parameters_weight_,
+            l_self_modules_features_modules_16_modules_branch2_modules_2_modules_bn_parameters_bias_,
+            False,
+            0.1,
+            0.001,
+        )
+        x_297 = l_self_modules_features_modules_16_modules_branch2_modules_2_modules_bn_buffers_running_mean_ = l_self_modules_features_modules_16_modules_branch2_modules_2_modules_bn_buffers_running_var_ = l_self_modules_features_modules_16_modules_branch2_modules_2_modules_bn_parameters_weight_ = l_self_modules_features_modules_16_modules_branch2_modules_2_modules_bn_parameters_bias_ = (None)
+        x_299 = torch.nn.functional.relu(x_298, inplace=True)
+        x_298 = None
+        x_300 = torch.conv2d(
+            x_299,
+            l_self_modules_features_modules_16_modules_branch2_modules_3_modules_conv_parameters_weight_,
+            None,
+            (1, 1),
+            (3, 0),
+            (1, 1),
+            1,
+        )
+        x_299 = l_self_modules_features_modules_16_modules_branch2_modules_3_modules_conv_parameters_weight_ = (None)
+        x_301 = torch.nn.functional.batch_norm(
+            x_300,
+            l_self_modules_features_modules_16_modules_branch2_modules_3_modules_bn_buffers_running_mean_,
+            l_self_modules_features_modules_16_modules_branch2_modules_3_modules_bn_buffers_running_var_,
+            l_self_modules_features_modules_16_modules_branch2_modules_3_modules_bn_parameters_weight_,
+            l_self_modules_features_modules_16_modules_branch2_modules_3_modules_bn_parameters_bias_,
+            False,
+            0.1,
+            0.001,
+        )
+        x_300 = l_self_modules_features_modules_16_modules_branch2_modules_3_modules_bn_buffers_running_mean_ = l_self_modules_features_modules_16_modules_branch2_modules_3_modules_bn_buffers_running_var_ = l_self_modules_features_modules_16_modules_branch2_modules_3_modules_bn_parameters_weight_ = l_self_modules_features_modules_16_modules_branch2_modules_3_modules_bn_parameters_bias_ = (None)
+        x_302 = torch.nn.functional.relu(x_301, inplace=True)
+        x_301 = None
+        x_303 = torch.conv2d(
+            x_302,
+            l_self_modules_features_modules_16_modules_branch2_modules_4_modules_conv_parameters_weight_,
+            None,
+            (1, 1),
+            (0, 3),
+            (1, 1),
+            1,
+        )
+        x_302 = l_self_modules_features_modules_16_modules_branch2_modules_4_modules_conv_parameters_weight_ = (None)
+        x_304 = torch.nn.functional.batch_norm(
+            x_303,
+            l_self_modules_features_modules_16_modules_branch2_modules_4_modules_bn_buffers_running_mean_,
+            l_self_modules_features_modules_16_modules_branch2_modules_4_modules_bn_buffers_running_var_,
+            l_self_modules_features_modules_16_modules_branch2_modules_4_modules_bn_parameters_weight_,
+            l_self_modules_features_modules_16_modules_branch2_modules_4_modules_bn_parameters_bias_,
+            False,
+            0.1,
+            0.001,
+        )
+        x_303 = l_self_modules_features_modules_16_modules_branch2_modules_4_modules_bn_buffers_running_mean_ = l_self_modules_features_modules_16_modules_branch2_modules_4_modules_bn_buffers_running_var_ = l_self_modules_features_modules_16_modules_branch2_modules_4_modules_bn_parameters_weight_ = l_self_modules_features_modules_16_modules_branch2_modules_4_modules_bn_parameters_bias_ = (None)
+        x_305 = torch.nn.functional.relu(x_304, inplace=True)
+        x_304 = None
+        input_10 = torch._C._nn.avg_pool2d(out_12, 3, 1, 1, False, False, None)
+        out_12 = None
+        x_306 = torch.conv2d(
+            input_10,
+            l_self_modules_features_modules_16_modules_branch3_modules_1_modules_conv_parameters_weight_,
+            None,
+            (1, 1),
+            (0, 0),
+            (1, 1),
+            1,
+        )
+        input_10 = l_self_modules_features_modules_16_modules_branch3_modules_1_modules_conv_parameters_weight_ = (None)
+        x_307 = torch.nn.functional.batch_norm(
+            x_306,
+            l_self_modules_features_modules_16_modules_branch3_modules_1_modules_bn_buffers_running_mean_,
+            l_self_modules_features_modules_16_modules_branch3_modules_1_modules_bn_buffers_running_var_,
+            l_self_modules_features_modules_16_modules_branch3_modules_1_modules_bn_parameters_weight_,
+            l_self_modules_features_modules_16_modules_branch3_modules_1_modules_bn_parameters_bias_,
+            False,
+            0.1,
+            0.001,
+        )
+        x_306 = l_self_modules_features_modules_16_modules_branch3_modules_1_modules_bn_buffers_running_mean_ = l_self_modules_features_modules_16_modules_branch3_modules_1_modules_bn_buffers_running_var_ = l_self_modules_features_modules_16_modules_branch3_modules_1_modules_bn_parameters_weight_ = l_self_modules_features_modules_16_modules_branch3_modules_1_modules_bn_parameters_bias_ = (None)
+        x_308 = torch.nn.functional.relu(x_307, inplace=True)
+        x_307 = None
+        out_13 = torch.cat((x_281, x_290, x_305, x_308), 1)
+        x_281 = x_290 = x_305 = x_308 = None
+        x_309 = torch.conv2d(
+            out_13,
+            l_self_modules_features_modules_17_modules_branch0_modules_conv_parameters_weight_,
+            None,
+            (1, 1),
+            (0, 0),
+            (1, 1),
+            1,
+        )
+        l_self_modules_features_modules_17_modules_branch0_modules_conv_parameters_weight_ = (
+            None
+        )
+        x_310 = torch.nn.functional.batch_norm(
+            x_309,
+            l_self_modules_features_modules_17_modules_branch0_modules_bn_buffers_running_mean_,
+            l_self_modules_features_modules_17_modules_branch0_modules_bn_buffers_running_var_,
+            l_self_modules_features_modules_17_modules_branch0_modules_bn_parameters_weight_,
+            l_self_modules_features_modules_17_modules_branch0_modules_bn_parameters_bias_,
+            False,
+            0.1,
+            0.001,
+        )
+        x_309 = l_self_modules_features_modules_17_modules_branch0_modules_bn_buffers_running_mean_ = l_self_modules_features_modules_17_modules_branch0_modules_bn_buffers_running_var_ = l_self_modules_features_modules_17_modules_branch0_modules_bn_parameters_weight_ = l_self_modules_features_modules_17_modules_branch0_modules_bn_parameters_bias_ = (None)
+        x_311 = torch.nn.functional.relu(x_310, inplace=True)
+        x_310 = None
+        x_312 = torch.conv2d(
+            out_13,
+            l_self_modules_features_modules_17_modules_branch1_modules_0_modules_conv_parameters_weight_,
+            None,
+            (1, 1),
+            (0, 0),
+            (1, 1),
+            1,
+        )
+        l_self_modules_features_modules_17_modules_branch1_modules_0_modules_conv_parameters_weight_ = (
+            None
+        )
+        x_313 = torch.nn.functional.batch_norm(
+            x_312,
+            l_self_modules_features_modules_17_modules_branch1_modules_0_modules_bn_buffers_running_mean_,
+            l_self_modules_features_modules_17_modules_branch1_modules_0_modules_bn_buffers_running_var_,
+            l_self_modules_features_modules_17_modules_branch1_modules_0_modules_bn_parameters_weight_,
+            l_self_modules_features_modules_17_modules_branch1_modules_0_modules_bn_parameters_bias_,
+            False,
+            0.1,
+            0.001,
+        )
+        x_312 = l_self_modules_features_modules_17_modules_branch1_modules_0_modules_bn_buffers_running_mean_ = l_self_modules_features_modules_17_modules_branch1_modules_0_modules_bn_buffers_running_var_ = l_self_modules_features_modules_17_modules_branch1_modules_0_modules_bn_parameters_weight_ = l_self_modules_features_modules_17_modules_branch1_modules_0_modules_bn_parameters_bias_ = (None)
+        x_314 = torch.nn.functional.relu(x_313, inplace=True)
+        x_313 = None
+        x_315 = torch.conv2d(
+            x_314,
+            l_self_modules_features_modules_17_modules_branch1_modules_1_modules_conv_parameters_weight_,
+            None,
+            (1, 1),
+            (0, 3),
+            (1, 1),
+            1,
+        )
+        x_314 = l_self_modules_features_modules_17_modules_branch1_modules_1_modules_conv_parameters_weight_ = (None)
+        x_316 = torch.nn.functional.batch_norm(
+            x_315,
+            l_self_modules_features_modules_17_modules_branch1_modules_1_modules_bn_buffers_running_mean_,
+            l_self_modules_features_modules_17_modules_branch1_modules_1_modules_bn_buffers_running_var_,
+            l_self_modules_features_modules_17_modules_branch1_modules_1_modules_bn_parameters_weight_,
+            l_self_modules_features_modules_17_modules_branch1_modules_1_modules_bn_parameters_bias_,
+            False,
+            0.1,
+            0.001,
+        )
+        x_315 = l_self_modules_features_modules_17_modules_branch1_modules_1_modules_bn_buffers_running_mean_ = l_self_modules_features_modules_17_modules_branch1_modules_1_modules_bn_buffers_running_var_ = l_self_modules_features_modules_17_modules_branch1_modules_1_modules_bn_parameters_weight_ = l_self_modules_features_modules_17_modules_branch1_modules_1_modules_bn_parameters_bias_ = (None)
+        x_317 = torch.nn.functional.relu(x_316, inplace=True)
+        x_316 = None
+        x_318 = torch.conv2d(
+            x_317,
+            l_self_modules_features_modules_17_modules_branch1_modules_2_modules_conv_parameters_weight_,
+            None,
+            (1, 1),
+            (3, 0),
+            (1, 1),
+            1,
+        )
+        x_317 = l_self_modules_features_modules_17_modules_branch1_modules_2_modules_conv_parameters_weight_ = (None)
+        x_319 = torch.nn.functional.batch_norm(
+            x_318,
+            l_self_modules_features_modules_17_modules_branch1_modules_2_modules_bn_buffers_running_mean_,
+            l_self_modules_features_modules_17_modules_branch1_modules_2_modules_bn_buffers_running_var_,
+            l_self_modules_features_modules_17_modules_branch1_modules_2_modules_bn_parameters_weight_,
+            l_self_modules_features_modules_17_modules_branch1_modules_2_modules_bn_parameters_bias_,
+            False,
+            0.1,
+            0.001,
+        )
+        x_318 = l_self_modules_features_modules_17_modules_branch1_modules_2_modules_bn_buffers_running_mean_ = l_self_modules_features_modules_17_modules_branch1_modules_2_modules_bn_buffers_running_var_ = l_self_modules_features_modules_17_modules_branch1_modules_2_modules_bn_parameters_weight_ = l_self_modules_features_modules_17_modules_branch1_modules_2_modules_bn_parameters_bias_ = (None)
+        x_320 = torch.nn.functional.relu(x_319, inplace=True)
+        x_319 = None
+        x_321 = torch.conv2d(
+            out_13,
+            l_self_modules_features_modules_17_modules_branch2_modules_0_modules_conv_parameters_weight_,
+            None,
+            (1, 1),
+            (0, 0),
+            (1, 1),
+            1,
+        )
+        l_self_modules_features_modules_17_modules_branch2_modules_0_modules_conv_parameters_weight_ = (
+            None
+        )
+        x_322 = torch.nn.functional.batch_norm(
+            x_321,
+            l_self_modules_features_modules_17_modules_branch2_modules_0_modules_bn_buffers_running_mean_,
+            l_self_modules_features_modules_17_modules_branch2_modules_0_modules_bn_buffers_running_var_,
+            l_self_modules_features_modules_17_modules_branch2_modules_0_modules_bn_parameters_weight_,
+            l_self_modules_features_modules_17_modules_branch2_modules_0_modules_bn_parameters_bias_,
+            False,
+            0.1,
+            0.001,
+        )
+        x_321 = l_self_modules_features_modules_17_modules_branch2_modules_0_modules_bn_buffers_running_mean_ = l_self_modules_features_modules_17_modules_branch2_modules_0_modules_bn_buffers_running_var_ = l_self_modules_features_modules_17_modules_branch2_modules_0_modules_bn_parameters_weight_ = l_self_modules_features_modules_17_modules_branch2_modules_0_modules_bn_parameters_bias_ = (None)
+        x_323 = torch.nn.functional.relu(x_322, inplace=True)
+        x_322 = None
+        x_324 = torch.conv2d(
+            x_323,
+            l_self_modules_features_modules_17_modules_branch2_modules_1_modules_conv_parameters_weight_,
+            None,
+            (1, 1),
+            (3, 0),
+            (1, 1),
+            1,
+        )
+        x_323 = l_self_modules_features_modules_17_modules_branch2_modules_1_modules_conv_parameters_weight_ = (None)
+        x_325 = torch.nn.functional.batch_norm(
+            x_324,
+            l_self_modules_features_modules_17_modules_branch2_modules_1_modules_bn_buffers_running_mean_,
+            l_self_modules_features_modules_17_modules_branch2_modules_1_modules_bn_buffers_running_var_,
+            l_self_modules_features_modules_17_modules_branch2_modules_1_modules_bn_parameters_weight_,
+            l_self_modules_features_modules_17_modules_branch2_modules_1_modules_bn_parameters_bias_,
+            False,
+            0.1,
+            0.001,
+        )
+        x_324 = l_self_modules_features_modules_17_modules_branch2_modules_1_modules_bn_buffers_running_mean_ = l_self_modules_features_modules_17_modules_branch2_modules_1_modules_bn_buffers_running_var_ = l_self_modules_features_modules_17_modules_branch2_modules_1_modules_bn_parameters_weight_ = l_self_modules_features_modules_17_modules_branch2_modules_1_modules_bn_parameters_bias_ = (None)
+        x_326 = torch.nn.functional.relu(x_325, inplace=True)
+        x_325 = None
+        x_327 = torch.conv2d(
+            x_326,
+            l_self_modules_features_modules_17_modules_branch2_modules_2_modules_conv_parameters_weight_,
+            None,
+            (1, 1),
+            (0, 3),
+            (1, 1),
+            1,
+        )
+        x_326 = l_self_modules_features_modules_17_modules_branch2_modules_2_modules_conv_parameters_weight_ = (None)
+        x_328 = torch.nn.functional.batch_norm(
+            x_327,
+            l_self_modules_features_modules_17_modules_branch2_modules_2_modules_bn_buffers_running_mean_,
+            l_self_modules_features_modules_17_modules_branch2_modules_2_modules_bn_buffers_running_var_,
+            l_self_modules_features_modules_17_modules_branch2_modules_2_modules_bn_parameters_weight_,
+            l_self_modules_features_modules_17_modules_branch2_modules_2_modules_bn_parameters_bias_,
+            False,
+            0.1,
+            0.001,
+        )
+        x_327 = l_self_modules_features_modules_17_modules_branch2_modules_2_modules_bn_buffers_running_mean_ = l_self_modules_features_modules_17_modules_branch2_modules_2_modules_bn_buffers_running_var_ = l_self_modules_features_modules_17_modules_branch2_modules_2_modules_bn_parameters_weight_ = l_self_modules_features_modules_17_modules_branch2_modules_2_modules_bn_parameters_bias_ = (None)
+        x_329 = torch.nn.functional.relu(x_328, inplace=True)
+        x_328 = None
+        x_330 = torch.conv2d(
+            x_329,
+            l_self_modules_features_modules_17_modules_branch2_modules_3_modules_conv_parameters_weight_,
+            None,
+            (1, 1),
+            (3, 0),
+            (1, 1),
+            1,
+        )
+        x_329 = l_self_modules_features_modules_17_modules_branch2_modules_3_modules_conv_parameters_weight_ = (None)
+        x_331 = torch.nn.functional.batch_norm(
+            x_330,
+            l_self_modules_features_modules_17_modules_branch2_modules_3_modules_bn_buffers_running_mean_,
+            l_self_modules_features_modules_17_modules_branch2_modules_3_modules_bn_buffers_running_var_,
+            l_self_modules_features_modules_17_modules_branch2_modules_3_modules_bn_parameters_weight_,
+            l_self_modules_features_modules_17_modules_branch2_modules_3_modules_bn_parameters_bias_,
+            False,
+            0.1,
+            0.001,
+        )
+        x_330 = l_self_modules_features_modules_17_modules_branch2_modules_3_modules_bn_buffers_running_mean_ = l_self_modules_features_modules_17_modules_branch2_modules_3_modules_bn_buffers_running_var_ = l_self_modules_features_modules_17_modules_branch2_modules_3_modules_bn_parameters_weight_ = l_self_modules_features_modules_17_modules_branch2_modules_3_modules_bn_parameters_bias_ = (None)
+        x_332 = torch.nn.functional.relu(x_331, inplace=True)
+        x_331 = None
+        x_333 = torch.conv2d(
+            x_332,
+            l_self_modules_features_modules_17_modules_branch2_modules_4_modules_conv_parameters_weight_,
+            None,
+            (1, 1),
+            (0, 3),
+            (1, 1),
+            1,
+        )
+        x_332 = l_self_modules_features_modules_17_modules_branch2_modules_4_modules_conv_parameters_weight_ = (None)
+        x_334 = torch.nn.functional.batch_norm(
+            x_333,
+            l_self_modules_features_modules_17_modules_branch2_modules_4_modules_bn_buffers_running_mean_,
+            l_self_modules_features_modules_17_modules_branch2_modules_4_modules_bn_buffers_running_var_,
+            l_self_modules_features_modules_17_modules_branch2_modules_4_modules_bn_parameters_weight_,
+            l_self_modules_features_modules_17_modules_branch2_modules_4_modules_bn_parameters_bias_,
+            False,
+            0.1,
+            0.001,
+        )
+        x_333 = l_self_modules_features_modules_17_modules_branch2_modules_4_modules_bn_buffers_running_mean_ = l_self_modules_features_modules_17_modules_branch2_modules_4_modules_bn_buffers_running_var_ = l_self_modules_features_modules_17_modules_branch2_modules_4_modules_bn_parameters_weight_ = l_self_modules_features_modules_17_modules_branch2_modules_4_modules_bn_parameters_bias_ = (None)
+        x_335 = torch.nn.functional.relu(x_334, inplace=True)
+        x_334 = None
+        input_11 = torch._C._nn.avg_pool2d(out_13, 3, 1, 1, False, False, None)
+        out_13 = None
+        x_336 = torch.conv2d(
+            input_11,
+            l_self_modules_features_modules_17_modules_branch3_modules_1_modules_conv_parameters_weight_,
+            None,
+            (1, 1),
+            (0, 0),
+            (1, 1),
+            1,
+        )
+        input_11 = l_self_modules_features_modules_17_modules_branch3_modules_1_modules_conv_parameters_weight_ = (None)
+        x_337 = torch.nn.functional.batch_norm(
+            x_336,
+            l_self_modules_features_modules_17_modules_branch3_modules_1_modules_bn_buffers_running_mean_,
+            l_self_modules_features_modules_17_modules_branch3_modules_1_modules_bn_buffers_running_var_,
+            l_self_modules_features_modules_17_modules_branch3_modules_1_modules_bn_parameters_weight_,
+            l_self_modules_features_modules_17_modules_branch3_modules_1_modules_bn_parameters_bias_,
+            False,
+            0.1,
+            0.001,
+        )
+        x_336 = l_self_modules_features_modules_17_modules_branch3_modules_1_modules_bn_buffers_running_mean_ = l_self_modules_features_modules_17_modules_branch3_modules_1_modules_bn_buffers_running_var_ = l_self_modules_features_modules_17_modules_branch3_modules_1_modules_bn_parameters_weight_ = l_self_modules_features_modules_17_modules_branch3_modules_1_modules_bn_parameters_bias_ = (None)
+        x_338 = torch.nn.functional.relu(x_337, inplace=True)
+        x_337 = None
+        out_14 = torch.cat((x_311, x_320, x_335, x_338), 1)
+        x_311 = x_320 = x_335 = x_338 = None
+        x_339 = torch.conv2d(
+            out_14,
+            l_self_modules_features_modules_18_modules_branch0_modules_0_modules_conv_parameters_weight_,
+            None,
+            (1, 1),
+            (0, 0),
+            (1, 1),
+            1,
+        )
+        l_self_modules_features_modules_18_modules_branch0_modules_0_modules_conv_parameters_weight_ = (
+            None
+        )
+        x_340 = torch.nn.functional.batch_norm(
+            x_339,
+            l_self_modules_features_modules_18_modules_branch0_modules_0_modules_bn_buffers_running_mean_,
+            l_self_modules_features_modules_18_modules_branch0_modules_0_modules_bn_buffers_running_var_,
+            l_self_modules_features_modules_18_modules_branch0_modules_0_modules_bn_parameters_weight_,
+            l_self_modules_features_modules_18_modules_branch0_modules_0_modules_bn_parameters_bias_,
+            False,
+            0.1,
+            0.001,
+        )
+        x_339 = l_self_modules_features_modules_18_modules_branch0_modules_0_modules_bn_buffers_running_mean_ = l_self_modules_features_modules_18_modules_branch0_modules_0_modules_bn_buffers_running_var_ = l_self_modules_features_modules_18_modules_branch0_modules_0_modules_bn_parameters_weight_ = l_self_modules_features_modules_18_modules_branch0_modules_0_modules_bn_parameters_bias_ = (None)
+        x_341 = torch.nn.functional.relu(x_340, inplace=True)
+        x_340 = None
+        x_342 = torch.conv2d(
+            x_341,
+            l_self_modules_features_modules_18_modules_branch0_modules_1_modules_conv_parameters_weight_,
+            None,
+            (2, 2),
+            (0, 0),
+            (1, 1),
+            1,
+        )
+        x_341 = l_self_modules_features_modules_18_modules_branch0_modules_1_modules_conv_parameters_weight_ = (None)
+        x_343 = torch.nn.functional.batch_norm(
+            x_342,
+            l_self_modules_features_modules_18_modules_branch0_modules_1_modules_bn_buffers_running_mean_,
+            l_self_modules_features_modules_18_modules_branch0_modules_1_modules_bn_buffers_running_var_,
+            l_self_modules_features_modules_18_modules_branch0_modules_1_modules_bn_parameters_weight_,
+            l_self_modules_features_modules_18_modules_branch0_modules_1_modules_bn_parameters_bias_,
+            False,
+            0.1,
+            0.001,
+        )
+        x_342 = l_self_modules_features_modules_18_modules_branch0_modules_1_modules_bn_buffers_running_mean_ = l_self_modules_features_modules_18_modules_branch0_modules_1_modules_bn_buffers_running_var_ = l_self_modules_features_modules_18_modules_branch0_modules_1_modules_bn_parameters_weight_ = l_self_modules_features_modules_18_modules_branch0_modules_1_modules_bn_parameters_bias_ = (None)
+        x_344 = torch.nn.functional.relu(x_343, inplace=True)
+        x_343 = None
+        x_345 = torch.conv2d(
+            out_14,
+            l_self_modules_features_modules_18_modules_branch1_modules_0_modules_conv_parameters_weight_,
+            None,
+            (1, 1),
+            (0, 0),
+            (1, 1),
+            1,
+        )
+        l_self_modules_features_modules_18_modules_branch1_modules_0_modules_conv_parameters_weight_ = (
+            None
+        )
+        x_346 = torch.nn.functional.batch_norm(
+            x_345,
+            l_self_modules_features_modules_18_modules_branch1_modules_0_modules_bn_buffers_running_mean_,
+            l_self_modules_features_modules_18_modules_branch1_modules_0_modules_bn_buffers_running_var_,
+            l_self_modules_features_modules_18_modules_branch1_modules_0_modules_bn_parameters_weight_,
+            l_self_modules_features_modules_18_modules_branch1_modules_0_modules_bn_parameters_bias_,
+            False,
+            0.1,
+            0.001,
+        )
+        x_345 = l_self_modules_features_modules_18_modules_branch1_modules_0_modules_bn_buffers_running_mean_ = l_self_modules_features_modules_18_modules_branch1_modules_0_modules_bn_buffers_running_var_ = l_self_modules_features_modules_18_modules_branch1_modules_0_modules_bn_parameters_weight_ = l_self_modules_features_modules_18_modules_branch1_modules_0_modules_bn_parameters_bias_ = (None)
+        x_347 = torch.nn.functional.relu(x_346, inplace=True)
+        x_346 = None
+        x_348 = torch.conv2d(
+            x_347,
+            l_self_modules_features_modules_18_modules_branch1_modules_1_modules_conv_parameters_weight_,
+            None,
+            (1, 1),
+            (0, 3),
+            (1, 1),
+            1,
+        )
+        x_347 = l_self_modules_features_modules_18_modules_branch1_modules_1_modules_conv_parameters_weight_ = (None)
+        x_349 = torch.nn.functional.batch_norm(
+            x_348,
+            l_self_modules_features_modules_18_modules_branch1_modules_1_modules_bn_buffers_running_mean_,
+            l_self_modules_features_modules_18_modules_branch1_modules_1_modules_bn_buffers_running_var_,
+            l_self_modules_features_modules_18_modules_branch1_modules_1_modules_bn_parameters_weight_,
+            l_self_modules_features_modules_18_modules_branch1_modules_1_modules_bn_parameters_bias_,
+            False,
+            0.1,
+            0.001,
+        )
+        x_348 = l_self_modules_features_modules_18_modules_branch1_modules_1_modules_bn_buffers_running_mean_ = l_self_modules_features_modules_18_modules_branch1_modules_1_modules_bn_buffers_running_var_ = l_self_modules_features_modules_18_modules_branch1_modules_1_modules_bn_parameters_weight_ = l_self_modules_features_modules_18_modules_branch1_modules_1_modules_bn_parameters_bias_ = (None)
+        x_350 = torch.nn.functional.relu(x_349, inplace=True)
+        x_349 = None
+        x_351 = torch.conv2d(
+            x_350,
+            l_self_modules_features_modules_18_modules_branch1_modules_2_modules_conv_parameters_weight_,
+            None,
+            (1, 1),
+            (3, 0),
+            (1, 1),
+            1,
+        )
+        x_350 = l_self_modules_features_modules_18_modules_branch1_modules_2_modules_conv_parameters_weight_ = (None)
+        x_352 = torch.nn.functional.batch_norm(
+            x_351,
+            l_self_modules_features_modules_18_modules_branch1_modules_2_modules_bn_buffers_running_mean_,
+            l_self_modules_features_modules_18_modules_branch1_modules_2_modules_bn_buffers_running_var_,
+            l_self_modules_features_modules_18_modules_branch1_modules_2_modules_bn_parameters_weight_,
+            l_self_modules_features_modules_18_modules_branch1_modules_2_modules_bn_parameters_bias_,
+            False,
+            0.1,
+            0.001,
+        )
+        x_351 = l_self_modules_features_modules_18_modules_branch1_modules_2_modules_bn_buffers_running_mean_ = l_self_modules_features_modules_18_modules_branch1_modules_2_modules_bn_buffers_running_var_ = l_self_modules_features_modules_18_modules_branch1_modules_2_modules_bn_parameters_weight_ = l_self_modules_features_modules_18_modules_branch1_modules_2_modules_bn_parameters_bias_ = (None)
+        x_353 = torch.nn.functional.relu(x_352, inplace=True)
+        x_352 = None
+        x_354 = torch.conv2d(
+            x_353,
+            l_self_modules_features_modules_18_modules_branch1_modules_3_modules_conv_parameters_weight_,
+            None,
+            (2, 2),
+            (0, 0),
+            (1, 1),
+            1,
+        )
+        x_353 = l_self_modules_features_modules_18_modules_branch1_modules_3_modules_conv_parameters_weight_ = (None)
+        x_355 = torch.nn.functional.batch_norm(
+            x_354,
+            l_self_modules_features_modules_18_modules_branch1_modules_3_modules_bn_buffers_running_mean_,
+            l_self_modules_features_modules_18_modules_branch1_modules_3_modules_bn_buffers_running_var_,
+            l_self_modules_features_modules_18_modules_branch1_modules_3_modules_bn_parameters_weight_,
+            l_self_modules_features_modules_18_modules_branch1_modules_3_modules_bn_parameters_bias_,
+            False,
+            0.1,
+            0.001,
+        )
+        x_354 = l_self_modules_features_modules_18_modules_branch1_modules_3_modules_bn_buffers_running_mean_ = l_self_modules_features_modules_18_modules_branch1_modules_3_modules_bn_buffers_running_var_ = l_self_modules_features_modules_18_modules_branch1_modules_3_modules_bn_parameters_weight_ = l_self_modules_features_modules_18_modules_branch1_modules_3_modules_bn_parameters_bias_ = (None)
+        x_356 = torch.nn.functional.relu(x_355, inplace=True)
+        x_355 = None
+        x2_1 = torch.nn.functional.max_pool2d(
+            out_14, 3, 2, 0, 1, ceil_mode=False, return_indices=False
+        )
+        out_14 = None
+        out_15 = torch.cat((x_344, x_356, x2_1), 1)
+        x_344 = x_356 = x2_1 = None
+        x_357 = torch.conv2d(
+            out_15,
+            l_self_modules_features_modules_19_modules_branch0_modules_conv_parameters_weight_,
+            None,
+            (1, 1),
+            (0, 0),
+            (1, 1),
+            1,
+        )
+        l_self_modules_features_modules_19_modules_branch0_modules_conv_parameters_weight_ = (
+            None
+        )
+        x_358 = torch.nn.functional.batch_norm(
+            x_357,
+            l_self_modules_features_modules_19_modules_branch0_modules_bn_buffers_running_mean_,
+            l_self_modules_features_modules_19_modules_branch0_modules_bn_buffers_running_var_,
+            l_self_modules_features_modules_19_modules_branch0_modules_bn_parameters_weight_,
+            l_self_modules_features_modules_19_modules_branch0_modules_bn_parameters_bias_,
+            False,
+            0.1,
+            0.001,
+        )
+        x_357 = l_self_modules_features_modules_19_modules_branch0_modules_bn_buffers_running_mean_ = l_self_modules_features_modules_19_modules_branch0_modules_bn_buffers_running_var_ = l_self_modules_features_modules_19_modules_branch0_modules_bn_parameters_weight_ = l_self_modules_features_modules_19_modules_branch0_modules_bn_parameters_bias_ = (None)
+        x_359 = torch.nn.functional.relu(x_358, inplace=True)
+        x_358 = None
+        x_360 = torch.conv2d(
+            out_15,
+            l_self_modules_features_modules_19_modules_branch1_0_modules_conv_parameters_weight_,
+            None,
+            (1, 1),
+            (0, 0),
+            (1, 1),
+            1,
+        )
+        l_self_modules_features_modules_19_modules_branch1_0_modules_conv_parameters_weight_ = (
+            None
+        )
+        x_361 = torch.nn.functional.batch_norm(
+            x_360,
+            l_self_modules_features_modules_19_modules_branch1_0_modules_bn_buffers_running_mean_,
+            l_self_modules_features_modules_19_modules_branch1_0_modules_bn_buffers_running_var_,
+            l_self_modules_features_modules_19_modules_branch1_0_modules_bn_parameters_weight_,
+            l_self_modules_features_modules_19_modules_branch1_0_modules_bn_parameters_bias_,
+            False,
+            0.1,
+            0.001,
+        )
+        x_360 = l_self_modules_features_modules_19_modules_branch1_0_modules_bn_buffers_running_mean_ = l_self_modules_features_modules_19_modules_branch1_0_modules_bn_buffers_running_var_ = l_self_modules_features_modules_19_modules_branch1_0_modules_bn_parameters_weight_ = l_self_modules_features_modules_19_modules_branch1_0_modules_bn_parameters_bias_ = (None)
+        x_362 = torch.nn.functional.relu(x_361, inplace=True)
+        x_361 = None
+        x_363 = torch.conv2d(
+            x_362,
+            l_self_modules_features_modules_19_modules_branch1_1a_modules_conv_parameters_weight_,
+            None,
+            (1, 1),
+            (0, 1),
+            (1, 1),
+            1,
+        )
+        l_self_modules_features_modules_19_modules_branch1_1a_modules_conv_parameters_weight_ = (
+            None
+        )
+        x_364 = torch.nn.functional.batch_norm(
+            x_363,
+            l_self_modules_features_modules_19_modules_branch1_1a_modules_bn_buffers_running_mean_,
+            l_self_modules_features_modules_19_modules_branch1_1a_modules_bn_buffers_running_var_,
+            l_self_modules_features_modules_19_modules_branch1_1a_modules_bn_parameters_weight_,
+            l_self_modules_features_modules_19_modules_branch1_1a_modules_bn_parameters_bias_,
+            False,
+            0.1,
+            0.001,
+        )
+        x_363 = l_self_modules_features_modules_19_modules_branch1_1a_modules_bn_buffers_running_mean_ = l_self_modules_features_modules_19_modules_branch1_1a_modules_bn_buffers_running_var_ = l_self_modules_features_modules_19_modules_branch1_1a_modules_bn_parameters_weight_ = l_self_modules_features_modules_19_modules_branch1_1a_modules_bn_parameters_bias_ = (None)
+        x_365 = torch.nn.functional.relu(x_364, inplace=True)
+        x_364 = None
+        x_366 = torch.conv2d(
+            x_362,
+            l_self_modules_features_modules_19_modules_branch1_1b_modules_conv_parameters_weight_,
+            None,
+            (1, 1),
+            (1, 0),
+            (1, 1),
+            1,
+        )
+        x_362 = l_self_modules_features_modules_19_modules_branch1_1b_modules_conv_parameters_weight_ = (None)
+        x_367 = torch.nn.functional.batch_norm(
+            x_366,
+            l_self_modules_features_modules_19_modules_branch1_1b_modules_bn_buffers_running_mean_,
+            l_self_modules_features_modules_19_modules_branch1_1b_modules_bn_buffers_running_var_,
+            l_self_modules_features_modules_19_modules_branch1_1b_modules_bn_parameters_weight_,
+            l_self_modules_features_modules_19_modules_branch1_1b_modules_bn_parameters_bias_,
+            False,
+            0.1,
+            0.001,
+        )
+        x_366 = l_self_modules_features_modules_19_modules_branch1_1b_modules_bn_buffers_running_mean_ = l_self_modules_features_modules_19_modules_branch1_1b_modules_bn_buffers_running_var_ = l_self_modules_features_modules_19_modules_branch1_1b_modules_bn_parameters_weight_ = l_self_modules_features_modules_19_modules_branch1_1b_modules_bn_parameters_bias_ = (None)
+        x_368 = torch.nn.functional.relu(x_367, inplace=True)
+        x_367 = None
+        x1_1 = torch.cat((x_365, x_368), 1)
+        x_365 = x_368 = None
+        x_369 = torch.conv2d(
+            out_15,
+            l_self_modules_features_modules_19_modules_branch2_0_modules_conv_parameters_weight_,
+            None,
+            (1, 1),
+            (0, 0),
+            (1, 1),
+            1,
+        )
+        l_self_modules_features_modules_19_modules_branch2_0_modules_conv_parameters_weight_ = (
+            None
+        )
+        x_370 = torch.nn.functional.batch_norm(
+            x_369,
+            l_self_modules_features_modules_19_modules_branch2_0_modules_bn_buffers_running_mean_,
+            l_self_modules_features_modules_19_modules_branch2_0_modules_bn_buffers_running_var_,
+            l_self_modules_features_modules_19_modules_branch2_0_modules_bn_parameters_weight_,
+            l_self_modules_features_modules_19_modules_branch2_0_modules_bn_parameters_bias_,
+            False,
+            0.1,
+            0.001,
+        )
+        x_369 = l_self_modules_features_modules_19_modules_branch2_0_modules_bn_buffers_running_mean_ = l_self_modules_features_modules_19_modules_branch2_0_modules_bn_buffers_running_var_ = l_self_modules_features_modules_19_modules_branch2_0_modules_bn_parameters_weight_ = l_self_modules_features_modules_19_modules_branch2_0_modules_bn_parameters_bias_ = (None)
+        x_371 = torch.nn.functional.relu(x_370, inplace=True)
+        x_370 = None
+        x_372 = torch.conv2d(
+            x_371,
+            l_self_modules_features_modules_19_modules_branch2_1_modules_conv_parameters_weight_,
+            None,
+            (1, 1),
+            (1, 0),
+            (1, 1),
+            1,
+        )
+        x_371 = l_self_modules_features_modules_19_modules_branch2_1_modules_conv_parameters_weight_ = (None)
+        x_373 = torch.nn.functional.batch_norm(
+            x_372,
+            l_self_modules_features_modules_19_modules_branch2_1_modules_bn_buffers_running_mean_,
+            l_self_modules_features_modules_19_modules_branch2_1_modules_bn_buffers_running_var_,
+            l_self_modules_features_modules_19_modules_branch2_1_modules_bn_parameters_weight_,
+            l_self_modules_features_modules_19_modules_branch2_1_modules_bn_parameters_bias_,
+            False,
+            0.1,
+            0.001,
+        )
+        x_372 = l_self_modules_features_modules_19_modules_branch2_1_modules_bn_buffers_running_mean_ = l_self_modules_features_modules_19_modules_branch2_1_modules_bn_buffers_running_var_ = l_self_modules_features_modules_19_modules_branch2_1_modules_bn_parameters_weight_ = l_self_modules_features_modules_19_modules_branch2_1_modules_bn_parameters_bias_ = (None)
+        x_374 = torch.nn.functional.relu(x_373, inplace=True)
+        x_373 = None
+        x_375 = torch.conv2d(
+            x_374,
+            l_self_modules_features_modules_19_modules_branch2_2_modules_conv_parameters_weight_,
+            None,
+            (1, 1),
+            (0, 1),
+            (1, 1),
+            1,
+        )
+        x_374 = l_self_modules_features_modules_19_modules_branch2_2_modules_conv_parameters_weight_ = (None)
+        x_376 = torch.nn.functional.batch_norm(
+            x_375,
+            l_self_modules_features_modules_19_modules_branch2_2_modules_bn_buffers_running_mean_,
+            l_self_modules_features_modules_19_modules_branch2_2_modules_bn_buffers_running_var_,
+            l_self_modules_features_modules_19_modules_branch2_2_modules_bn_parameters_weight_,
+            l_self_modules_features_modules_19_modules_branch2_2_modules_bn_parameters_bias_,
+            False,
+            0.1,
+            0.001,
+        )
+        x_375 = l_self_modules_features_modules_19_modules_branch2_2_modules_bn_buffers_running_mean_ = l_self_modules_features_modules_19_modules_branch2_2_modules_bn_buffers_running_var_ = l_self_modules_features_modules_19_modules_branch2_2_modules_bn_parameters_weight_ = l_self_modules_features_modules_19_modules_branch2_2_modules_bn_parameters_bias_ = (None)
+        x_377 = torch.nn.functional.relu(x_376, inplace=True)
+        x_376 = None
+        x_378 = torch.conv2d(
+            x_377,
+            l_self_modules_features_modules_19_modules_branch2_3a_modules_conv_parameters_weight_,
+            None,
+            (1, 1),
+            (0, 1),
+            (1, 1),
+            1,
+        )
+        l_self_modules_features_modules_19_modules_branch2_3a_modules_conv_parameters_weight_ = (
+            None
+        )
+        x_379 = torch.nn.functional.batch_norm(
+            x_378,
+            l_self_modules_features_modules_19_modules_branch2_3a_modules_bn_buffers_running_mean_,
+            l_self_modules_features_modules_19_modules_branch2_3a_modules_bn_buffers_running_var_,
+            l_self_modules_features_modules_19_modules_branch2_3a_modules_bn_parameters_weight_,
+            l_self_modules_features_modules_19_modules_branch2_3a_modules_bn_parameters_bias_,
+            False,
+            0.1,
+            0.001,
+        )
+        x_378 = l_self_modules_features_modules_19_modules_branch2_3a_modules_bn_buffers_running_mean_ = l_self_modules_features_modules_19_modules_branch2_3a_modules_bn_buffers_running_var_ = l_self_modules_features_modules_19_modules_branch2_3a_modules_bn_parameters_weight_ = l_self_modules_features_modules_19_modules_branch2_3a_modules_bn_parameters_bias_ = (None)
+        x_380 = torch.nn.functional.relu(x_379, inplace=True)
+        x_379 = None
+        x_381 = torch.conv2d(
+            x_377,
+            l_self_modules_features_modules_19_modules_branch2_3b_modules_conv_parameters_weight_,
+            None,
+            (1, 1),
+            (1, 0),
+            (1, 1),
+            1,
+        )
+        x_377 = l_self_modules_features_modules_19_modules_branch2_3b_modules_conv_parameters_weight_ = (None)
+        x_382 = torch.nn.functional.batch_norm(
+            x_381,
+            l_self_modules_features_modules_19_modules_branch2_3b_modules_bn_buffers_running_mean_,
+            l_self_modules_features_modules_19_modules_branch2_3b_modules_bn_buffers_running_var_,
+            l_self_modules_features_modules_19_modules_branch2_3b_modules_bn_parameters_weight_,
+            l_self_modules_features_modules_19_modules_branch2_3b_modules_bn_parameters_bias_,
+            False,
+            0.1,
+            0.001,
+        )
+        x_381 = l_self_modules_features_modules_19_modules_branch2_3b_modules_bn_buffers_running_mean_ = l_self_modules_features_modules_19_modules_branch2_3b_modules_bn_buffers_running_var_ = l_self_modules_features_modules_19_modules_branch2_3b_modules_bn_parameters_weight_ = l_self_modules_features_modules_19_modules_branch2_3b_modules_bn_parameters_bias_ = (None)
+        x_383 = torch.nn.functional.relu(x_382, inplace=True)
+        x_382 = None
+        x2_2 = torch.cat((x_380, x_383), 1)
+        x_380 = x_383 = None
+        input_12 = torch._C._nn.avg_pool2d(out_15, 3, 1, 1, False, False, None)
+        out_15 = None
+        x_384 = torch.conv2d(
+            input_12,
+            l_self_modules_features_modules_19_modules_branch3_modules_1_modules_conv_parameters_weight_,
+            None,
+            (1, 1),
+            (0, 0),
+            (1, 1),
+            1,
+        )
+        input_12 = l_self_modules_features_modules_19_modules_branch3_modules_1_modules_conv_parameters_weight_ = (None)
+        x_385 = torch.nn.functional.batch_norm(
+            x_384,
+            l_self_modules_features_modules_19_modules_branch3_modules_1_modules_bn_buffers_running_mean_,
+            l_self_modules_features_modules_19_modules_branch3_modules_1_modules_bn_buffers_running_var_,
+            l_self_modules_features_modules_19_modules_branch3_modules_1_modules_bn_parameters_weight_,
+            l_self_modules_features_modules_19_modules_branch3_modules_1_modules_bn_parameters_bias_,
+            False,
+            0.1,
+            0.001,
+        )
+        x_384 = l_self_modules_features_modules_19_modules_branch3_modules_1_modules_bn_buffers_running_mean_ = l_self_modules_features_modules_19_modules_branch3_modules_1_modules_bn_buffers_running_var_ = l_self_modules_features_modules_19_modules_branch3_modules_1_modules_bn_parameters_weight_ = l_self_modules_features_modules_19_modules_branch3_modules_1_modules_bn_parameters_bias_ = (None)
+        x_386 = torch.nn.functional.relu(x_385, inplace=True)
+        x_385 = None
+        out_16 = torch.cat((x_359, x1_1, x2_2, x_386), 1)
+        x_359 = x1_1 = x2_2 = x_386 = None
+        x_387 = torch.conv2d(
+            out_16,
+            l_self_modules_features_modules_20_modules_branch0_modules_conv_parameters_weight_,
+            None,
+            (1, 1),
+            (0, 0),
+            (1, 1),
+            1,
+        )
+        l_self_modules_features_modules_20_modules_branch0_modules_conv_parameters_weight_ = (
+            None
+        )
+        x_388 = torch.nn.functional.batch_norm(
+            x_387,
+            l_self_modules_features_modules_20_modules_branch0_modules_bn_buffers_running_mean_,
+            l_self_modules_features_modules_20_modules_branch0_modules_bn_buffers_running_var_,
+            l_self_modules_features_modules_20_modules_branch0_modules_bn_parameters_weight_,
+            l_self_modules_features_modules_20_modules_branch0_modules_bn_parameters_bias_,
+            False,
+            0.1,
+            0.001,
+        )
+        x_387 = l_self_modules_features_modules_20_modules_branch0_modules_bn_buffers_running_mean_ = l_self_modules_features_modules_20_modules_branch0_modules_bn_buffers_running_var_ = l_self_modules_features_modules_20_modules_branch0_modules_bn_parameters_weight_ = l_self_modules_features_modules_20_modules_branch0_modules_bn_parameters_bias_ = (None)
+        x_389 = torch.nn.functional.relu(x_388, inplace=True)
+        x_388 = None
+        x_390 = torch.conv2d(
+            out_16,
+            l_self_modules_features_modules_20_modules_branch1_0_modules_conv_parameters_weight_,
+            None,
+            (1, 1),
+            (0, 0),
+            (1, 1),
+            1,
+        )
+        l_self_modules_features_modules_20_modules_branch1_0_modules_conv_parameters_weight_ = (
+            None
+        )
+        x_391 = torch.nn.functional.batch_norm(
+            x_390,
+            l_self_modules_features_modules_20_modules_branch1_0_modules_bn_buffers_running_mean_,
+            l_self_modules_features_modules_20_modules_branch1_0_modules_bn_buffers_running_var_,
+            l_self_modules_features_modules_20_modules_branch1_0_modules_bn_parameters_weight_,
+            l_self_modules_features_modules_20_modules_branch1_0_modules_bn_parameters_bias_,
+            False,
+            0.1,
+            0.001,
+        )
+        x_390 = l_self_modules_features_modules_20_modules_branch1_0_modules_bn_buffers_running_mean_ = l_self_modules_features_modules_20_modules_branch1_0_modules_bn_buffers_running_var_ = l_self_modules_features_modules_20_modules_branch1_0_modules_bn_parameters_weight_ = l_self_modules_features_modules_20_modules_branch1_0_modules_bn_parameters_bias_ = (None)
+        x_392 = torch.nn.functional.relu(x_391, inplace=True)
+        x_391 = None
+        x_393 = torch.conv2d(
+            x_392,
+            l_self_modules_features_modules_20_modules_branch1_1a_modules_conv_parameters_weight_,
+            None,
+            (1, 1),
+            (0, 1),
+            (1, 1),
+            1,
+        )
+        l_self_modules_features_modules_20_modules_branch1_1a_modules_conv_parameters_weight_ = (
+            None
+        )
+        x_394 = torch.nn.functional.batch_norm(
+            x_393,
+            l_self_modules_features_modules_20_modules_branch1_1a_modules_bn_buffers_running_mean_,
+            l_self_modules_features_modules_20_modules_branch1_1a_modules_bn_buffers_running_var_,
+            l_self_modules_features_modules_20_modules_branch1_1a_modules_bn_parameters_weight_,
+            l_self_modules_features_modules_20_modules_branch1_1a_modules_bn_parameters_bias_,
+            False,
+            0.1,
+            0.001,
+        )
+        x_393 = l_self_modules_features_modules_20_modules_branch1_1a_modules_bn_buffers_running_mean_ = l_self_modules_features_modules_20_modules_branch1_1a_modules_bn_buffers_running_var_ = l_self_modules_features_modules_20_modules_branch1_1a_modules_bn_parameters_weight_ = l_self_modules_features_modules_20_modules_branch1_1a_modules_bn_parameters_bias_ = (None)
+        x_395 = torch.nn.functional.relu(x_394, inplace=True)
+        x_394 = None
+        x_396 = torch.conv2d(
+            x_392,
+            l_self_modules_features_modules_20_modules_branch1_1b_modules_conv_parameters_weight_,
+            None,
+            (1, 1),
+            (1, 0),
+            (1, 1),
+            1,
+        )
+        x_392 = l_self_modules_features_modules_20_modules_branch1_1b_modules_conv_parameters_weight_ = (None)
+        x_397 = torch.nn.functional.batch_norm(
+            x_396,
+            l_self_modules_features_modules_20_modules_branch1_1b_modules_bn_buffers_running_mean_,
+            l_self_modules_features_modules_20_modules_branch1_1b_modules_bn_buffers_running_var_,
+            l_self_modules_features_modules_20_modules_branch1_1b_modules_bn_parameters_weight_,
+            l_self_modules_features_modules_20_modules_branch1_1b_modules_bn_parameters_bias_,
+            False,
+            0.1,
+            0.001,
+        )
+        x_396 = l_self_modules_features_modules_20_modules_branch1_1b_modules_bn_buffers_running_mean_ = l_self_modules_features_modules_20_modules_branch1_1b_modules_bn_buffers_running_var_ = l_self_modules_features_modules_20_modules_branch1_1b_modules_bn_parameters_weight_ = l_self_modules_features_modules_20_modules_branch1_1b_modules_bn_parameters_bias_ = (None)
+        x_398 = torch.nn.functional.relu(x_397, inplace=True)
+        x_397 = None
+        x1_2 = torch.cat((x_395, x_398), 1)
+        x_395 = x_398 = None
+        x_399 = torch.conv2d(
+            out_16,
+            l_self_modules_features_modules_20_modules_branch2_0_modules_conv_parameters_weight_,
+            None,
+            (1, 1),
+            (0, 0),
+            (1, 1),
+            1,
+        )
+        l_self_modules_features_modules_20_modules_branch2_0_modules_conv_parameters_weight_ = (
+            None
+        )
+        x_400 = torch.nn.functional.batch_norm(
+            x_399,
+            l_self_modules_features_modules_20_modules_branch2_0_modules_bn_buffers_running_mean_,
+            l_self_modules_features_modules_20_modules_branch2_0_modules_bn_buffers_running_var_,
+            l_self_modules_features_modules_20_modules_branch2_0_modules_bn_parameters_weight_,
+            l_self_modules_features_modules_20_modules_branch2_0_modules_bn_parameters_bias_,
+            False,
+            0.1,
+            0.001,
+        )
+        x_399 = l_self_modules_features_modules_20_modules_branch2_0_modules_bn_buffers_running_mean_ = l_self_modules_features_modules_20_modules_branch2_0_modules_bn_buffers_running_var_ = l_self_modules_features_modules_20_modules_branch2_0_modules_bn_parameters_weight_ = l_self_modules_features_modules_20_modules_branch2_0_modules_bn_parameters_bias_ = (None)
+        x_401 = torch.nn.functional.relu(x_400, inplace=True)
+        x_400 = None
+        x_402 = torch.conv2d(
+            x_401,
+            l_self_modules_features_modules_20_modules_branch2_1_modules_conv_parameters_weight_,
+            None,
+            (1, 1),
+            (1, 0),
+            (1, 1),
+            1,
+        )
+        x_401 = l_self_modules_features_modules_20_modules_branch2_1_modules_conv_parameters_weight_ = (None)
+        x_403 = torch.nn.functional.batch_norm(
+            x_402,
+            l_self_modules_features_modules_20_modules_branch2_1_modules_bn_buffers_running_mean_,
+            l_self_modules_features_modules_20_modules_branch2_1_modules_bn_buffers_running_var_,
+            l_self_modules_features_modules_20_modules_branch2_1_modules_bn_parameters_weight_,
+            l_self_modules_features_modules_20_modules_branch2_1_modules_bn_parameters_bias_,
+            False,
+            0.1,
+            0.001,
+        )
+        x_402 = l_self_modules_features_modules_20_modules_branch2_1_modules_bn_buffers_running_mean_ = l_self_modules_features_modules_20_modules_branch2_1_modules_bn_buffers_running_var_ = l_self_modules_features_modules_20_modules_branch2_1_modules_bn_parameters_weight_ = l_self_modules_features_modules_20_modules_branch2_1_modules_bn_parameters_bias_ = (None)
+        x_404 = torch.nn.functional.relu(x_403, inplace=True)
+        x_403 = None
+        x_405 = torch.conv2d(
+            x_404,
+            l_self_modules_features_modules_20_modules_branch2_2_modules_conv_parameters_weight_,
+            None,
+            (1, 1),
+            (0, 1),
+            (1, 1),
+            1,
+        )
+        x_404 = l_self_modules_features_modules_20_modules_branch2_2_modules_conv_parameters_weight_ = (None)
+        x_406 = torch.nn.functional.batch_norm(
+            x_405,
+            l_self_modules_features_modules_20_modules_branch2_2_modules_bn_buffers_running_mean_,
+            l_self_modules_features_modules_20_modules_branch2_2_modules_bn_buffers_running_var_,
+            l_self_modules_features_modules_20_modules_branch2_2_modules_bn_parameters_weight_,
+            l_self_modules_features_modules_20_modules_branch2_2_modules_bn_parameters_bias_,
+            False,
+            0.1,
+            0.001,
+        )
+        x_405 = l_self_modules_features_modules_20_modules_branch2_2_modules_bn_buffers_running_mean_ = l_self_modules_features_modules_20_modules_branch2_2_modules_bn_buffers_running_var_ = l_self_modules_features_modules_20_modules_branch2_2_modules_bn_parameters_weight_ = l_self_modules_features_modules_20_modules_branch2_2_modules_bn_parameters_bias_ = (None)
+        x_407 = torch.nn.functional.relu(x_406, inplace=True)
+        x_406 = None
+        x_408 = torch.conv2d(
+            x_407,
+            l_self_modules_features_modules_20_modules_branch2_3a_modules_conv_parameters_weight_,
+            None,
+            (1, 1),
+            (0, 1),
+            (1, 1),
+            1,
+        )
+        l_self_modules_features_modules_20_modules_branch2_3a_modules_conv_parameters_weight_ = (
+            None
+        )
+        x_409 = torch.nn.functional.batch_norm(
+            x_408,
+            l_self_modules_features_modules_20_modules_branch2_3a_modules_bn_buffers_running_mean_,
+            l_self_modules_features_modules_20_modules_branch2_3a_modules_bn_buffers_running_var_,
+            l_self_modules_features_modules_20_modules_branch2_3a_modules_bn_parameters_weight_,
+            l_self_modules_features_modules_20_modules_branch2_3a_modules_bn_parameters_bias_,
+            False,
+            0.1,
+            0.001,
+        )
+        x_408 = l_self_modules_features_modules_20_modules_branch2_3a_modules_bn_buffers_running_mean_ = l_self_modules_features_modules_20_modules_branch2_3a_modules_bn_buffers_running_var_ = l_self_modules_features_modules_20_modules_branch2_3a_modules_bn_parameters_weight_ = l_self_modules_features_modules_20_modules_branch2_3a_modules_bn_parameters_bias_ = (None)
+        x_410 = torch.nn.functional.relu(x_409, inplace=True)
+        x_409 = None
+        x_411 = torch.conv2d(
+            x_407,
+            l_self_modules_features_modules_20_modules_branch2_3b_modules_conv_parameters_weight_,
+            None,
+            (1, 1),
+            (1, 0),
+            (1, 1),
+            1,
+        )
+        x_407 = l_self_modules_features_modules_20_modules_branch2_3b_modules_conv_parameters_weight_ = (None)
+        x_412 = torch.nn.functional.batch_norm(
+            x_411,
+            l_self_modules_features_modules_20_modules_branch2_3b_modules_bn_buffers_running_mean_,
+            l_self_modules_features_modules_20_modules_branch2_3b_modules_bn_buffers_running_var_,
+            l_self_modules_features_modules_20_modules_branch2_3b_modules_bn_parameters_weight_,
+            l_self_modules_features_modules_20_modules_branch2_3b_modules_bn_parameters_bias_,
+            False,
+            0.1,
+            0.001,
+        )
+        x_411 = l_self_modules_features_modules_20_modules_branch2_3b_modules_bn_buffers_running_mean_ = l_self_modules_features_modules_20_modules_branch2_3b_modules_bn_buffers_running_var_ = l_self_modules_features_modules_20_modules_branch2_3b_modules_bn_parameters_weight_ = l_self_modules_features_modules_20_modules_branch2_3b_modules_bn_parameters_bias_ = (None)
+        x_413 = torch.nn.functional.relu(x_412, inplace=True)
+        x_412 = None
+        x2_3 = torch.cat((x_410, x_413), 1)
+        x_410 = x_413 = None
+        input_13 = torch._C._nn.avg_pool2d(out_16, 3, 1, 1, False, False, None)
+        out_16 = None
+        x_414 = torch.conv2d(
+            input_13,
+            l_self_modules_features_modules_20_modules_branch3_modules_1_modules_conv_parameters_weight_,
+            None,
+            (1, 1),
+            (0, 0),
+            (1, 1),
+            1,
+        )
+        input_13 = l_self_modules_features_modules_20_modules_branch3_modules_1_modules_conv_parameters_weight_ = (None)
+        x_415 = torch.nn.functional.batch_norm(
+            x_414,
+            l_self_modules_features_modules_20_modules_branch3_modules_1_modules_bn_buffers_running_mean_,
+            l_self_modules_features_modules_20_modules_branch3_modules_1_modules_bn_buffers_running_var_,
+            l_self_modules_features_modules_20_modules_branch3_modules_1_modules_bn_parameters_weight_,
+            l_self_modules_features_modules_20_modules_branch3_modules_1_modules_bn_parameters_bias_,
+            False,
+            0.1,
+            0.001,
+        )
+        x_414 = l_self_modules_features_modules_20_modules_branch3_modules_1_modules_bn_buffers_running_mean_ = l_self_modules_features_modules_20_modules_branch3_modules_1_modules_bn_buffers_running_var_ = l_self_modules_features_modules_20_modules_branch3_modules_1_modules_bn_parameters_weight_ = l_self_modules_features_modules_20_modules_branch3_modules_1_modules_bn_parameters_bias_ = (None)
+        x_416 = torch.nn.functional.relu(x_415, inplace=True)
+        x_415 = None
+        out_17 = torch.cat((x_389, x1_2, x2_3, x_416), 1)
+        x_389 = x1_2 = x2_3 = x_416 = None
+        x_417 = torch.conv2d(
+            out_17,
+            l_self_modules_features_modules_21_modules_branch0_modules_conv_parameters_weight_,
+            None,
+            (1, 1),
+            (0, 0),
+            (1, 1),
+            1,
+        )
+        l_self_modules_features_modules_21_modules_branch0_modules_conv_parameters_weight_ = (
+            None
+        )
+        x_418 = torch.nn.functional.batch_norm(
+            x_417,
+            l_self_modules_features_modules_21_modules_branch0_modules_bn_buffers_running_mean_,
+            l_self_modules_features_modules_21_modules_branch0_modules_bn_buffers_running_var_,
+            l_self_modules_features_modules_21_modules_branch0_modules_bn_parameters_weight_,
+            l_self_modules_features_modules_21_modules_branch0_modules_bn_parameters_bias_,
+            False,
+            0.1,
+            0.001,
+        )
+        x_417 = l_self_modules_features_modules_21_modules_branch0_modules_bn_buffers_running_mean_ = l_self_modules_features_modules_21_modules_branch0_modules_bn_buffers_running_var_ = l_self_modules_features_modules_21_modules_branch0_modules_bn_parameters_weight_ = l_self_modules_features_modules_21_modules_branch0_modules_bn_parameters_bias_ = (None)
+        x_419 = torch.nn.functional.relu(x_418, inplace=True)
+        x_418 = None
+        x_420 = torch.conv2d(
+            out_17,
+            l_self_modules_features_modules_21_modules_branch1_0_modules_conv_parameters_weight_,
+            None,
+            (1, 1),
+            (0, 0),
+            (1, 1),
+            1,
+        )
+        l_self_modules_features_modules_21_modules_branch1_0_modules_conv_parameters_weight_ = (
+            None
+        )
+        x_421 = torch.nn.functional.batch_norm(
+            x_420,
+            l_self_modules_features_modules_21_modules_branch1_0_modules_bn_buffers_running_mean_,
+            l_self_modules_features_modules_21_modules_branch1_0_modules_bn_buffers_running_var_,
+            l_self_modules_features_modules_21_modules_branch1_0_modules_bn_parameters_weight_,
+            l_self_modules_features_modules_21_modules_branch1_0_modules_bn_parameters_bias_,
+            False,
+            0.1,
+            0.001,
+        )
+        x_420 = l_self_modules_features_modules_21_modules_branch1_0_modules_bn_buffers_running_mean_ = l_self_modules_features_modules_21_modules_branch1_0_modules_bn_buffers_running_var_ = l_self_modules_features_modules_21_modules_branch1_0_modules_bn_parameters_weight_ = l_self_modules_features_modules_21_modules_branch1_0_modules_bn_parameters_bias_ = (None)
+        x_422 = torch.nn.functional.relu(x_421, inplace=True)
+        x_421 = None
+        x_423 = torch.conv2d(
+            x_422,
+            l_self_modules_features_modules_21_modules_branch1_1a_modules_conv_parameters_weight_,
+            None,
+            (1, 1),
+            (0, 1),
+            (1, 1),
+            1,
+        )
+        l_self_modules_features_modules_21_modules_branch1_1a_modules_conv_parameters_weight_ = (
+            None
+        )
+        x_424 = torch.nn.functional.batch_norm(
+            x_423,
+            l_self_modules_features_modules_21_modules_branch1_1a_modules_bn_buffers_running_mean_,
+            l_self_modules_features_modules_21_modules_branch1_1a_modules_bn_buffers_running_var_,
+            l_self_modules_features_modules_21_modules_branch1_1a_modules_bn_parameters_weight_,
+            l_self_modules_features_modules_21_modules_branch1_1a_modules_bn_parameters_bias_,
+            False,
+            0.1,
+            0.001,
+        )
+        x_423 = l_self_modules_features_modules_21_modules_branch1_1a_modules_bn_buffers_running_mean_ = l_self_modules_features_modules_21_modules_branch1_1a_modules_bn_buffers_running_var_ = l_self_modules_features_modules_21_modules_branch1_1a_modules_bn_parameters_weight_ = l_self_modules_features_modules_21_modules_branch1_1a_modules_bn_parameters_bias_ = (None)
+        x_425 = torch.nn.functional.relu(x_424, inplace=True)
+        x_424 = None
+        x_426 = torch.conv2d(
+            x_422,
+            l_self_modules_features_modules_21_modules_branch1_1b_modules_conv_parameters_weight_,
+            None,
+            (1, 1),
+            (1, 0),
+            (1, 1),
+            1,
+        )
+        x_422 = l_self_modules_features_modules_21_modules_branch1_1b_modules_conv_parameters_weight_ = (None)
+        x_427 = torch.nn.functional.batch_norm(
+            x_426,
+            l_self_modules_features_modules_21_modules_branch1_1b_modules_bn_buffers_running_mean_,
+            l_self_modules_features_modules_21_modules_branch1_1b_modules_bn_buffers_running_var_,
+            l_self_modules_features_modules_21_modules_branch1_1b_modules_bn_parameters_weight_,
+            l_self_modules_features_modules_21_modules_branch1_1b_modules_bn_parameters_bias_,
+            False,
+            0.1,
+            0.001,
+        )
+        x_426 = l_self_modules_features_modules_21_modules_branch1_1b_modules_bn_buffers_running_mean_ = l_self_modules_features_modules_21_modules_branch1_1b_modules_bn_buffers_running_var_ = l_self_modules_features_modules_21_modules_branch1_1b_modules_bn_parameters_weight_ = l_self_modules_features_modules_21_modules_branch1_1b_modules_bn_parameters_bias_ = (None)
+        x_428 = torch.nn.functional.relu(x_427, inplace=True)
+        x_427 = None
+        x1_3 = torch.cat((x_425, x_428), 1)
+        x_425 = x_428 = None
+        x_429 = torch.conv2d(
+            out_17,
+            l_self_modules_features_modules_21_modules_branch2_0_modules_conv_parameters_weight_,
+            None,
+            (1, 1),
+            (0, 0),
+            (1, 1),
+            1,
+        )
+        l_self_modules_features_modules_21_modules_branch2_0_modules_conv_parameters_weight_ = (
+            None
+        )
+        x_430 = torch.nn.functional.batch_norm(
+            x_429,
+            l_self_modules_features_modules_21_modules_branch2_0_modules_bn_buffers_running_mean_,
+            l_self_modules_features_modules_21_modules_branch2_0_modules_bn_buffers_running_var_,
+            l_self_modules_features_modules_21_modules_branch2_0_modules_bn_parameters_weight_,
+            l_self_modules_features_modules_21_modules_branch2_0_modules_bn_parameters_bias_,
+            False,
+            0.1,
+            0.001,
+        )
+        x_429 = l_self_modules_features_modules_21_modules_branch2_0_modules_bn_buffers_running_mean_ = l_self_modules_features_modules_21_modules_branch2_0_modules_bn_buffers_running_var_ = l_self_modules_features_modules_21_modules_branch2_0_modules_bn_parameters_weight_ = l_self_modules_features_modules_21_modules_branch2_0_modules_bn_parameters_bias_ = (None)
+        x_431 = torch.nn.functional.relu(x_430, inplace=True)
+        x_430 = None
+        x_432 = torch.conv2d(
+            x_431,
+            l_self_modules_features_modules_21_modules_branch2_1_modules_conv_parameters_weight_,
+            None,
+            (1, 1),
+            (1, 0),
+            (1, 1),
+            1,
+        )
+        x_431 = l_self_modules_features_modules_21_modules_branch2_1_modules_conv_parameters_weight_ = (None)
+        x_433 = torch.nn.functional.batch_norm(
+            x_432,
+            l_self_modules_features_modules_21_modules_branch2_1_modules_bn_buffers_running_mean_,
+            l_self_modules_features_modules_21_modules_branch2_1_modules_bn_buffers_running_var_,
+            l_self_modules_features_modules_21_modules_branch2_1_modules_bn_parameters_weight_,
+            l_self_modules_features_modules_21_modules_branch2_1_modules_bn_parameters_bias_,
+            False,
+            0.1,
+            0.001,
+        )
+        x_432 = l_self_modules_features_modules_21_modules_branch2_1_modules_bn_buffers_running_mean_ = l_self_modules_features_modules_21_modules_branch2_1_modules_bn_buffers_running_var_ = l_self_modules_features_modules_21_modules_branch2_1_modules_bn_parameters_weight_ = l_self_modules_features_modules_21_modules_branch2_1_modules_bn_parameters_bias_ = (None)
+        x_434 = torch.nn.functional.relu(x_433, inplace=True)
+        x_433 = None
+        x_435 = torch.conv2d(
+            x_434,
+            l_self_modules_features_modules_21_modules_branch2_2_modules_conv_parameters_weight_,
+            None,
+            (1, 1),
+            (0, 1),
+            (1, 1),
+            1,
+        )
+        x_434 = l_self_modules_features_modules_21_modules_branch2_2_modules_conv_parameters_weight_ = (None)
+        x_436 = torch.nn.functional.batch_norm(
+            x_435,
+            l_self_modules_features_modules_21_modules_branch2_2_modules_bn_buffers_running_mean_,
+            l_self_modules_features_modules_21_modules_branch2_2_modules_bn_buffers_running_var_,
+            l_self_modules_features_modules_21_modules_branch2_2_modules_bn_parameters_weight_,
+            l_self_modules_features_modules_21_modules_branch2_2_modules_bn_parameters_bias_,
+            False,
+            0.1,
+            0.001,
+        )
+        x_435 = l_self_modules_features_modules_21_modules_branch2_2_modules_bn_buffers_running_mean_ = l_self_modules_features_modules_21_modules_branch2_2_modules_bn_buffers_running_var_ = l_self_modules_features_modules_21_modules_branch2_2_modules_bn_parameters_weight_ = l_self_modules_features_modules_21_modules_branch2_2_modules_bn_parameters_bias_ = (None)
+        x_437 = torch.nn.functional.relu(x_436, inplace=True)
+        x_436 = None
+        x_438 = torch.conv2d(
+            x_437,
+            l_self_modules_features_modules_21_modules_branch2_3a_modules_conv_parameters_weight_,
+            None,
+            (1, 1),
+            (0, 1),
+            (1, 1),
+            1,
+        )
+        l_self_modules_features_modules_21_modules_branch2_3a_modules_conv_parameters_weight_ = (
+            None
+        )
+        x_439 = torch.nn.functional.batch_norm(
+            x_438,
+            l_self_modules_features_modules_21_modules_branch2_3a_modules_bn_buffers_running_mean_,
+            l_self_modules_features_modules_21_modules_branch2_3a_modules_bn_buffers_running_var_,
+            l_self_modules_features_modules_21_modules_branch2_3a_modules_bn_parameters_weight_,
+            l_self_modules_features_modules_21_modules_branch2_3a_modules_bn_parameters_bias_,
+            False,
+            0.1,
+            0.001,
+        )
+        x_438 = l_self_modules_features_modules_21_modules_branch2_3a_modules_bn_buffers_running_mean_ = l_self_modules_features_modules_21_modules_branch2_3a_modules_bn_buffers_running_var_ = l_self_modules_features_modules_21_modules_branch2_3a_modules_bn_parameters_weight_ = l_self_modules_features_modules_21_modules_branch2_3a_modules_bn_parameters_bias_ = (None)
+        x_440 = torch.nn.functional.relu(x_439, inplace=True)
+        x_439 = None
+        x_441 = torch.conv2d(
+            x_437,
+            l_self_modules_features_modules_21_modules_branch2_3b_modules_conv_parameters_weight_,
+            None,
+            (1, 1),
+            (1, 0),
+            (1, 1),
+            1,
+        )
+        x_437 = l_self_modules_features_modules_21_modules_branch2_3b_modules_conv_parameters_weight_ = (None)
+        x_442 = torch.nn.functional.batch_norm(
+            x_441,
+            l_self_modules_features_modules_21_modules_branch2_3b_modules_bn_buffers_running_mean_,
+            l_self_modules_features_modules_21_modules_branch2_3b_modules_bn_buffers_running_var_,
+            l_self_modules_features_modules_21_modules_branch2_3b_modules_bn_parameters_weight_,
+            l_self_modules_features_modules_21_modules_branch2_3b_modules_bn_parameters_bias_,
+            False,
+            0.1,
+            0.001,
+        )
+        x_441 = l_self_modules_features_modules_21_modules_branch2_3b_modules_bn_buffers_running_mean_ = l_self_modules_features_modules_21_modules_branch2_3b_modules_bn_buffers_running_var_ = l_self_modules_features_modules_21_modules_branch2_3b_modules_bn_parameters_weight_ = l_self_modules_features_modules_21_modules_branch2_3b_modules_bn_parameters_bias_ = (None)
+        x_443 = torch.nn.functional.relu(x_442, inplace=True)
+        x_442 = None
+        x2_4 = torch.cat((x_440, x_443), 1)
+        x_440 = x_443 = None
+        input_14 = torch._C._nn.avg_pool2d(out_17, 3, 1, 1, False, False, None)
+        out_17 = None
+        x_444 = torch.conv2d(
+            input_14,
+            l_self_modules_features_modules_21_modules_branch3_modules_1_modules_conv_parameters_weight_,
+            None,
+            (1, 1),
+            (0, 0),
+            (1, 1),
+            1,
+        )
+        input_14 = l_self_modules_features_modules_21_modules_branch3_modules_1_modules_conv_parameters_weight_ = (None)
+        x_445 = torch.nn.functional.batch_norm(
+            x_444,
+            l_self_modules_features_modules_21_modules_branch3_modules_1_modules_bn_buffers_running_mean_,
+            l_self_modules_features_modules_21_modules_branch3_modules_1_modules_bn_buffers_running_var_,
+            l_self_modules_features_modules_21_modules_branch3_modules_1_modules_bn_parameters_weight_,
+            l_self_modules_features_modules_21_modules_branch3_modules_1_modules_bn_parameters_bias_,
+            False,
+            0.1,
+            0.001,
+        )
+        x_444 = l_self_modules_features_modules_21_modules_branch3_modules_1_modules_bn_buffers_running_mean_ = l_self_modules_features_modules_21_modules_branch3_modules_1_modules_bn_buffers_running_var_ = l_self_modules_features_modules_21_modules_branch3_modules_1_modules_bn_parameters_weight_ = l_self_modules_features_modules_21_modules_branch3_modules_1_modules_bn_parameters_bias_ = (None)
+        x_446 = torch.nn.functional.relu(x_445, inplace=True)
+        x_445 = None
+        out_18 = torch.cat((x_419, x1_3, x2_4, x_446), 1)
+        x_419 = x1_3 = x2_4 = x_446 = None
+        x_447 = torch.nn.functional.adaptive_avg_pool2d(out_18, 1)
+        out_18 = None
+        x_448 = x_447.flatten(1, -1)
+        x_447 = None
+        x_449 = torch.nn.functional.dropout(x_448, 0.0, False, False)
+        x_448 = None
+        x_450 = torch._C._nn.linear(
+            x_449,
+            l_self_modules_last_linear_parameters_weight_,
+            l_self_modules_last_linear_parameters_bias_,
+        )
+        x_449 = (
+            l_self_modules_last_linear_parameters_weight_
+        ) = l_self_modules_last_linear_parameters_bias_ = None
+        return (x_450,)
diff --git a/samples/timm/inception_v4/weight_meta.py b/samples/timm/inception_v4/weight_meta.py
new file mode 100644
index 000000000..50305c3dd
--- /dev/null
+++ b/samples/timm/inception_v4/weight_meta.py
@@ -0,0 +1,7530 @@
+class Program_weight_tensor_meta_L_self_modules_features_modules_0_modules_conv_parameters_weight_:
+    name = "L_self_modules_features_modules_0_modules_conv_parameters_weight_"
+    shape = [32, 3, 3, 3]
+    dtype = "torch.float32"
+    device = "cpu"
+    mean = -0.007
+    std = 0.274
+    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, 299, 299]
+    dtype = "torch.float32"
+    device = "cpu"
+    mean = 0.003
+    std = 1.000
+    data = None
+
+
+class Program_weight_tensor_meta_L_self_modules_features_modules_0_modules_bn_buffers_running_mean_:
+    name = "L_self_modules_features_modules_0_modules_bn_buffers_running_mean_"
+    shape = [32]
+    dtype = "torch.float32"
+    device = "cpu"
+    mean = 0.048
+    std = 0.150
+    data = None
+
+
+class Program_weight_tensor_meta_L_self_modules_features_modules_0_modules_bn_buffers_running_var_:
+    name = "L_self_modules_features_modules_0_modules_bn_buffers_running_var_"
+    shape = [32]
+    dtype = "torch.float32"
+    device = "cpu"
+    mean = 0.498
+    std = 0.720
+    data = None
+
+
+class Program_weight_tensor_meta_L_self_modules_features_modules_0_modules_bn_parameters_weight_:
+    name = "L_self_modules_features_modules_0_modules_bn_parameters_weight_"
+    shape = [32]
+    dtype = "torch.float32"
+    device = "cpu"
+    mean = 1.000
+    std = 0.000
+    data = None
+
+
+class Program_weight_tensor_meta_L_self_modules_features_modules_0_modules_bn_parameters_bias_:
+    name = "L_self_modules_features_modules_0_modules_bn_parameters_bias_"
+    shape = [32]
+    dtype = "torch.float32"
+    device = "cpu"
+    mean = 0.052
+    std = 2.915
+    data = None
+
+
+class Program_weight_tensor_meta_L_self_modules_features_modules_1_modules_conv_parameters_weight_:
+    name = "L_self_modules_features_modules_1_modules_conv_parameters_weight_"
+    shape = [32, 32, 3, 3]
+    dtype = "torch.float32"
+    device = "cpu"
+    mean = 0.001
+    std = 0.140
+    data = None
+
+
+class Program_weight_tensor_meta_L_self_modules_features_modules_1_modules_bn_buffers_running_mean_:
+    name = "L_self_modules_features_modules_1_modules_bn_buffers_running_mean_"
+    shape = [32]
+    dtype = "torch.float32"
+    device = "cpu"
+    mean = -1.043
+    std = 4.996
+    data = None
+
+
+class Program_weight_tensor_meta_L_self_modules_features_modules_1_modules_bn_buffers_running_var_:
+    name = "L_self_modules_features_modules_1_modules_bn_buffers_running_var_"
+    shape = [32]
+    dtype = "torch.float32"
+    device = "cpu"
+    mean = 5.712
+    std = 6.823
+    data = None
+
+
+class Program_weight_tensor_meta_L_self_modules_features_modules_1_modules_bn_parameters_weight_:
+    name = "L_self_modules_features_modules_1_modules_bn_parameters_weight_"
+    shape = [32]
+    dtype = "torch.float32"
+    device = "cpu"
+    mean = 1.000
+    std = 0.000
+    data = None
+
+
+class Program_weight_tensor_meta_L_self_modules_features_modules_1_modules_bn_parameters_bias_:
+    name = "L_self_modules_features_modules_1_modules_bn_parameters_bias_"
+    shape = [32]
+    dtype = "torch.float32"
+    device = "cpu"
+    mean = -0.047
+    std = 1.714
+    data = None
+
+
+class Program_weight_tensor_meta_L_self_modules_features_modules_2_modules_conv_parameters_weight_:
+    name = "L_self_modules_features_modules_2_modules_conv_parameters_weight_"
+    shape = [64, 32, 3, 3]
+    dtype = "torch.float32"
+    device = "cpu"
+    mean = -0.007
+    std = 0.097
+    data = None
+
+
+class Program_weight_tensor_meta_L_self_modules_features_modules_2_modules_bn_buffers_running_mean_:
+    name = "L_self_modules_features_modules_2_modules_bn_buffers_running_mean_"
+    shape = [64]
+    dtype = "torch.float32"
+    device = "cpu"
+    mean = -1.810
+    std = 2.427
+    data = None
+
+
+class Program_weight_tensor_meta_L_self_modules_features_modules_2_modules_bn_buffers_running_var_:
+    name = "L_self_modules_features_modules_2_modules_bn_buffers_running_var_"
+    shape = [64]
+    dtype = "torch.float32"
+    device = "cpu"
+    mean = 3.588
+    std = 3.202
+    data = None
+
+
+class Program_weight_tensor_meta_L_self_modules_features_modules_2_modules_bn_parameters_weight_:
+    name = "L_self_modules_features_modules_2_modules_bn_parameters_weight_"
+    shape = [64]
+    dtype = "torch.float32"
+    device = "cpu"
+    mean = 1.000
+    std = 0.000
+    data = None
+
+
+class Program_weight_tensor_meta_L_self_modules_features_modules_2_modules_bn_parameters_bias_:
+    name = "L_self_modules_features_modules_2_modules_bn_parameters_bias_"
+    shape = [64]
+    dtype = "torch.float32"
+    device = "cpu"
+    mean = 0.239
+    std = 1.095
+    data = None
+
+
+class Program_weight_tensor_meta_L_self_modules_features_modules_3_modules_conv_modules_conv_parameters_weight_:
+    name = (
+        "L_self_modules_features_modules_3_modules_conv_modules_conv_parameters_weight_"
+    )
+    shape = [96, 64, 3, 3]
+    dtype = "torch.float32"
+    device = "cpu"
+    mean = -0.002
+    std = 0.050
+    data = None
+
+
+class Program_weight_tensor_meta_L_self_modules_features_modules_3_modules_conv_modules_bn_buffers_running_mean_:
+    name = "L_self_modules_features_modules_3_modules_conv_modules_bn_buffers_running_mean_"
+    shape = [96]
+    dtype = "torch.float32"
+    device = "cpu"
+    mean = -0.461
+    std = 3.167
+    data = None
+
+
+class Program_weight_tensor_meta_L_self_modules_features_modules_3_modules_conv_modules_bn_buffers_running_var_:
+    name = (
+        "L_self_modules_features_modules_3_modules_conv_modules_bn_buffers_running_var_"
+    )
+    shape = [96]
+    dtype = "torch.float32"
+    device = "cpu"
+    mean = 2.778
+    std = 1.022
+    data = None
+
+
+class Program_weight_tensor_meta_L_self_modules_features_modules_3_modules_conv_modules_bn_parameters_weight_:
+    name = (
+        "L_self_modules_features_modules_3_modules_conv_modules_bn_parameters_weight_"
+    )
+    shape = [96]
+    dtype = "torch.float32"
+    device = "cpu"
+    mean = 1.000
+    std = 0.000
+    data = None
+
+
+class Program_weight_tensor_meta_L_self_modules_features_modules_3_modules_conv_modules_bn_parameters_bias_:
+    name = "L_self_modules_features_modules_3_modules_conv_modules_bn_parameters_bias_"
+    shape = [96]
+    dtype = "torch.float32"
+    device = "cpu"
+    mean = -0.114
+    std = 0.418
+    data = None
+
+
+class Program_weight_tensor_meta_L_self_modules_features_modules_4_modules_branch0_modules_0_modules_conv_parameters_weight_:
+    name = "L_self_modules_features_modules_4_modules_branch0_modules_0_modules_conv_parameters_weight_"
+    shape = [64, 160, 1, 1]
+    dtype = "torch.float32"
+    device = "cpu"
+    mean = -0.004
+    std = 0.105
+    data = None
+
+
+class Program_weight_tensor_meta_L_self_modules_features_modules_4_modules_branch0_modules_0_modules_bn_buffers_running_mean_:
+    name = "L_self_modules_features_modules_4_modules_branch0_modules_0_modules_bn_buffers_running_mean_"
+    shape = [64]
+    dtype = "torch.float32"
+    device = "cpu"
+    mean = -0.531
+    std = 1.015
+    data = None
+
+
+class Program_weight_tensor_meta_L_self_modules_features_modules_4_modules_branch0_modules_0_modules_bn_buffers_running_var_:
+    name = "L_self_modules_features_modules_4_modules_branch0_modules_0_modules_bn_buffers_running_var_"
+    shape = [64]
+    dtype = "torch.float32"
+    device = "cpu"
+    mean = 1.355
+    std = 0.602
+    data = None
+
+
+class Program_weight_tensor_meta_L_self_modules_features_modules_4_modules_branch0_modules_0_modules_bn_parameters_weight_:
+    name = "L_self_modules_features_modules_4_modules_branch0_modules_0_modules_bn_parameters_weight_"
+    shape = [64]
+    dtype = "torch.float32"
+    device = "cpu"
+    mean = 1.000
+    std = 0.000
+    data = None
+
+
+class Program_weight_tensor_meta_L_self_modules_features_modules_4_modules_branch0_modules_0_modules_bn_parameters_bias_:
+    name = "L_self_modules_features_modules_4_modules_branch0_modules_0_modules_bn_parameters_bias_"
+    shape = [64]
+    dtype = "torch.float32"
+    device = "cpu"
+    mean = 0.071
+    std = 0.574
+    data = None
+
+
+class Program_weight_tensor_meta_L_self_modules_features_modules_4_modules_branch0_modules_1_modules_conv_parameters_weight_:
+    name = "L_self_modules_features_modules_4_modules_branch0_modules_1_modules_conv_parameters_weight_"
+    shape = [96, 64, 3, 3]
+    dtype = "torch.float32"
+    device = "cpu"
+    mean = -0.001
+    std = 0.056
+    data = None
+
+
+class Program_weight_tensor_meta_L_self_modules_features_modules_4_modules_branch0_modules_1_modules_bn_buffers_running_mean_:
+    name = "L_self_modules_features_modules_4_modules_branch0_modules_1_modules_bn_buffers_running_mean_"
+    shape = [96]
+    dtype = "torch.float32"
+    device = "cpu"
+    mean = -0.372
+    std = 0.932
+    data = None
+
+
+class Program_weight_tensor_meta_L_self_modules_features_modules_4_modules_branch0_modules_1_modules_bn_buffers_running_var_:
+    name = "L_self_modules_features_modules_4_modules_branch0_modules_1_modules_bn_buffers_running_var_"
+    shape = [96]
+    dtype = "torch.float32"
+    device = "cpu"
+    mean = 2.115
+    std = 1.107
+    data = None
+
+
+class Program_weight_tensor_meta_L_self_modules_features_modules_4_modules_branch0_modules_1_modules_bn_parameters_weight_:
+    name = "L_self_modules_features_modules_4_modules_branch0_modules_1_modules_bn_parameters_weight_"
+    shape = [96]
+    dtype = "torch.float32"
+    device = "cpu"
+    mean = 1.000
+    std = 0.000
+    data = None
+
+
+class Program_weight_tensor_meta_L_self_modules_features_modules_4_modules_branch0_modules_1_modules_bn_parameters_bias_:
+    name = "L_self_modules_features_modules_4_modules_branch0_modules_1_modules_bn_parameters_bias_"
+    shape = [96]
+    dtype = "torch.float32"
+    device = "cpu"
+    mean = 0.110
+    std = 1.212
+    data = None
+
+
+class Program_weight_tensor_meta_L_self_modules_features_modules_4_modules_branch1_modules_0_modules_conv_parameters_weight_:
+    name = "L_self_modules_features_modules_4_modules_branch1_modules_0_modules_conv_parameters_weight_"
+    shape = [64, 160, 1, 1]
+    dtype = "torch.float32"
+    device = "cpu"
+    mean = -0.004
+    std = 0.084
+    data = None
+
+
+class Program_weight_tensor_meta_L_self_modules_features_modules_4_modules_branch1_modules_0_modules_bn_buffers_running_mean_:
+    name = "L_self_modules_features_modules_4_modules_branch1_modules_0_modules_bn_buffers_running_mean_"
+    shape = [64]
+    dtype = "torch.float32"
+    device = "cpu"
+    mean = -0.616
+    std = 2.006
+    data = None
+
+
+class Program_weight_tensor_meta_L_self_modules_features_modules_4_modules_branch1_modules_0_modules_bn_buffers_running_var_:
+    name = "L_self_modules_features_modules_4_modules_branch1_modules_0_modules_bn_buffers_running_var_"
+    shape = [64]
+    dtype = "torch.float32"
+    device = "cpu"
+    mean = 1.223
+    std = 1.547
+    data = None
+
+
+class Program_weight_tensor_meta_L_self_modules_features_modules_4_modules_branch1_modules_0_modules_bn_parameters_weight_:
+    name = "L_self_modules_features_modules_4_modules_branch1_modules_0_modules_bn_parameters_weight_"
+    shape = [64]
+    dtype = "torch.float32"
+    device = "cpu"
+    mean = 1.000
+    std = 0.000
+    data = None
+
+
+class Program_weight_tensor_meta_L_self_modules_features_modules_4_modules_branch1_modules_0_modules_bn_parameters_bias_:
+    name = "L_self_modules_features_modules_4_modules_branch1_modules_0_modules_bn_parameters_bias_"
+    shape = [64]
+    dtype = "torch.float32"
+    device = "cpu"
+    mean = -0.350
+    std = 1.009
+    data = None
+
+
+class Program_weight_tensor_meta_L_self_modules_features_modules_4_modules_branch1_modules_1_modules_conv_parameters_weight_:
+    name = "L_self_modules_features_modules_4_modules_branch1_modules_1_modules_conv_parameters_weight_"
+    shape = [64, 64, 1, 7]
+    dtype = "torch.float32"
+    device = "cpu"
+    mean = -0.001
+    std = 0.057
+    data = None
+
+
+class Program_weight_tensor_meta_L_self_modules_features_modules_4_modules_branch1_modules_1_modules_bn_buffers_running_mean_:
+    name = "L_self_modules_features_modules_4_modules_branch1_modules_1_modules_bn_buffers_running_mean_"
+    shape = [64]
+    dtype = "torch.float32"
+    device = "cpu"
+    mean = -0.400
+    std = 1.124
+    data = None
+
+
+class Program_weight_tensor_meta_L_self_modules_features_modules_4_modules_branch1_modules_1_modules_bn_buffers_running_var_:
+    name = "L_self_modules_features_modules_4_modules_branch1_modules_1_modules_bn_buffers_running_var_"
+    shape = [64]
+    dtype = "torch.float32"
+    device = "cpu"
+    mean = 1.290
+    std = 1.146
+    data = None
+
+
+class Program_weight_tensor_meta_L_self_modules_features_modules_4_modules_branch1_modules_1_modules_bn_parameters_weight_:
+    name = "L_self_modules_features_modules_4_modules_branch1_modules_1_modules_bn_parameters_weight_"
+    shape = [64]
+    dtype = "torch.float32"
+    device = "cpu"
+    mean = 1.000
+    std = 0.000
+    data = None
+
+
+class Program_weight_tensor_meta_L_self_modules_features_modules_4_modules_branch1_modules_1_modules_bn_parameters_bias_:
+    name = "L_self_modules_features_modules_4_modules_branch1_modules_1_modules_bn_parameters_bias_"
+    shape = [64]
+    dtype = "torch.float32"
+    device = "cpu"
+    mean = 0.095
+    std = 1.006
+    data = None
+
+
+class Program_weight_tensor_meta_L_self_modules_features_modules_4_modules_branch1_modules_2_modules_conv_parameters_weight_:
+    name = "L_self_modules_features_modules_4_modules_branch1_modules_2_modules_conv_parameters_weight_"
+    shape = [64, 64, 7, 1]
+    dtype = "torch.float32"
+    device = "cpu"
+    mean = -0.002
+    std = 0.061
+    data = None
+
+
+class Program_weight_tensor_meta_L_self_modules_features_modules_4_modules_branch1_modules_2_modules_bn_buffers_running_mean_:
+    name = "L_self_modules_features_modules_4_modules_branch1_modules_2_modules_bn_buffers_running_mean_"
+    shape = [64]
+    dtype = "torch.float32"
+    device = "cpu"
+    mean = -0.977
+    std = 1.141
+    data = None
+
+
+class Program_weight_tensor_meta_L_self_modules_features_modules_4_modules_branch1_modules_2_modules_bn_buffers_running_var_:
+    name = "L_self_modules_features_modules_4_modules_branch1_modules_2_modules_bn_buffers_running_var_"
+    shape = [64]
+    dtype = "torch.float32"
+    device = "cpu"
+    mean = 1.615
+    std = 1.194
+    data = None
+
+
+class Program_weight_tensor_meta_L_self_modules_features_modules_4_modules_branch1_modules_2_modules_bn_parameters_weight_:
+    name = "L_self_modules_features_modules_4_modules_branch1_modules_2_modules_bn_parameters_weight_"
+    shape = [64]
+    dtype = "torch.float32"
+    device = "cpu"
+    mean = 1.000
+    std = 0.000
+    data = None
+
+
+class Program_weight_tensor_meta_L_self_modules_features_modules_4_modules_branch1_modules_2_modules_bn_parameters_bias_:
+    name = "L_self_modules_features_modules_4_modules_branch1_modules_2_modules_bn_parameters_bias_"
+    shape = [64]
+    dtype = "torch.float32"
+    device = "cpu"
+    mean = 0.049
+    std = 1.217
+    data = None
+
+
+class Program_weight_tensor_meta_L_self_modules_features_modules_4_modules_branch1_modules_3_modules_conv_parameters_weight_:
+    name = "L_self_modules_features_modules_4_modules_branch1_modules_3_modules_conv_parameters_weight_"
+    shape = [96, 64, 3, 3]
+    dtype = "torch.float32"
+    device = "cpu"
+    mean = -0.001
+    std = 0.050
+    data = None
+
+
+class Program_weight_tensor_meta_L_self_modules_features_modules_4_modules_branch1_modules_3_modules_bn_buffers_running_mean_:
+    name = "L_self_modules_features_modules_4_modules_branch1_modules_3_modules_bn_buffers_running_mean_"
+    shape = [96]
+    dtype = "torch.float32"
+    device = "cpu"
+    mean = -0.245
+    std = 0.881
+    data = None
+
+
+class Program_weight_tensor_meta_L_self_modules_features_modules_4_modules_branch1_modules_3_modules_bn_buffers_running_var_:
+    name = "L_self_modules_features_modules_4_modules_branch1_modules_3_modules_bn_buffers_running_var_"
+    shape = [96]
+    dtype = "torch.float32"
+    device = "cpu"
+    mean = 1.711
+    std = 0.592
+    data = None
+
+
+class Program_weight_tensor_meta_L_self_modules_features_modules_4_modules_branch1_modules_3_modules_bn_parameters_weight_:
+    name = "L_self_modules_features_modules_4_modules_branch1_modules_3_modules_bn_parameters_weight_"
+    shape = [96]
+    dtype = "torch.float32"
+    device = "cpu"
+    mean = 1.000
+    std = 0.000
+    data = None
+
+
+class Program_weight_tensor_meta_L_self_modules_features_modules_4_modules_branch1_modules_3_modules_bn_parameters_bias_:
+    name = "L_self_modules_features_modules_4_modules_branch1_modules_3_modules_bn_parameters_bias_"
+    shape = [96]
+    dtype = "torch.float32"
+    device = "cpu"
+    mean = -0.202
+    std = 0.678
+    data = None
+
+
+class Program_weight_tensor_meta_L_self_modules_features_modules_5_modules_conv_modules_conv_parameters_weight_:
+    name = (
+        "L_self_modules_features_modules_5_modules_conv_modules_conv_parameters_weight_"
+    )
+    shape = [192, 192, 3, 3]
+    dtype = "torch.float32"
+    device = "cpu"
+    mean = -0.000
+    std = 0.023
+    data = None
+
+
+class Program_weight_tensor_meta_L_self_modules_features_modules_5_modules_conv_modules_bn_buffers_running_mean_:
+    name = "L_self_modules_features_modules_5_modules_conv_modules_bn_buffers_running_mean_"
+    shape = [192]
+    dtype = "torch.float32"
+    device = "cpu"
+    mean = 0.128
+    std = 2.539
+    data = None
+
+
+class Program_weight_tensor_meta_L_self_modules_features_modules_5_modules_conv_modules_bn_buffers_running_var_:
+    name = (
+        "L_self_modules_features_modules_5_modules_conv_modules_bn_buffers_running_var_"
+    )
+    shape = [192]
+    dtype = "torch.float32"
+    device = "cpu"
+    mean = 1.564
+    std = 0.455
+    data = None
+
+
+class Program_weight_tensor_meta_L_self_modules_features_modules_5_modules_conv_modules_bn_parameters_weight_:
+    name = (
+        "L_self_modules_features_modules_5_modules_conv_modules_bn_parameters_weight_"
+    )
+    shape = [192]
+    dtype = "torch.float32"
+    device = "cpu"
+    mean = 1.000
+    std = 0.000
+    data = None
+
+
+class Program_weight_tensor_meta_L_self_modules_features_modules_5_modules_conv_modules_bn_parameters_bias_:
+    name = "L_self_modules_features_modules_5_modules_conv_modules_bn_parameters_bias_"
+    shape = [192]
+    dtype = "torch.float32"
+    device = "cpu"
+    mean = -0.262
+    std = 0.268
+    data = None
+
+
+class Program_weight_tensor_meta_L_self_modules_features_modules_6_modules_branch0_modules_conv_parameters_weight_:
+    name = "L_self_modules_features_modules_6_modules_branch0_modules_conv_parameters_weight_"
+    shape = [96, 384, 1, 1]
+    dtype = "torch.float32"
+    device = "cpu"
+    mean = -0.001
+    std = 0.064
+    data = None
+
+
+class Program_weight_tensor_meta_L_self_modules_features_modules_6_modules_branch0_modules_bn_buffers_running_mean_:
+    name = "L_self_modules_features_modules_6_modules_branch0_modules_bn_buffers_running_mean_"
+    shape = [96]
+    dtype = "torch.float32"
+    device = "cpu"
+    mean = -0.301
+    std = 2.276
+    data = None
+
+
+class Program_weight_tensor_meta_L_self_modules_features_modules_6_modules_branch0_modules_bn_buffers_running_var_:
+    name = "L_self_modules_features_modules_6_modules_branch0_modules_bn_buffers_running_var_"
+    shape = [96]
+    dtype = "torch.float32"
+    device = "cpu"
+    mean = 1.768
+    std = 1.404
+    data = None
+
+
+class Program_weight_tensor_meta_L_self_modules_features_modules_6_modules_branch0_modules_bn_parameters_weight_:
+    name = "L_self_modules_features_modules_6_modules_branch0_modules_bn_parameters_weight_"
+    shape = [96]
+    dtype = "torch.float32"
+    device = "cpu"
+    mean = 1.000
+    std = 0.000
+    data = None
+
+
+class Program_weight_tensor_meta_L_self_modules_features_modules_6_modules_branch0_modules_bn_parameters_bias_:
+    name = (
+        "L_self_modules_features_modules_6_modules_branch0_modules_bn_parameters_bias_"
+    )
+    shape = [96]
+    dtype = "torch.float32"
+    device = "cpu"
+    mean = 0.059
+    std = 0.429
+    data = None
+
+
+class Program_weight_tensor_meta_L_self_modules_features_modules_6_modules_branch1_modules_0_modules_conv_parameters_weight_:
+    name = "L_self_modules_features_modules_6_modules_branch1_modules_0_modules_conv_parameters_weight_"
+    shape = [64, 384, 1, 1]
+    dtype = "torch.float32"
+    device = "cpu"
+    mean = -0.002
+    std = 0.051
+    data = None
+
+
+class Program_weight_tensor_meta_L_self_modules_features_modules_6_modules_branch1_modules_0_modules_bn_buffers_running_mean_:
+    name = "L_self_modules_features_modules_6_modules_branch1_modules_0_modules_bn_buffers_running_mean_"
+    shape = [64]
+    dtype = "torch.float32"
+    device = "cpu"
+    mean = -0.145
+    std = 0.645
+    data = None
+
+
+class Program_weight_tensor_meta_L_self_modules_features_modules_6_modules_branch1_modules_0_modules_bn_buffers_running_var_:
+    name = "L_self_modules_features_modules_6_modules_branch1_modules_0_modules_bn_buffers_running_var_"
+    shape = [64]
+    dtype = "torch.float32"
+    device = "cpu"
+    mean = 1.072
+    std = 0.524
+    data = None
+
+
+class Program_weight_tensor_meta_L_self_modules_features_modules_6_modules_branch1_modules_0_modules_bn_parameters_weight_:
+    name = "L_self_modules_features_modules_6_modules_branch1_modules_0_modules_bn_parameters_weight_"
+    shape = [64]
+    dtype = "torch.float32"
+    device = "cpu"
+    mean = 1.000
+    std = 0.000
+    data = None
+
+
+class Program_weight_tensor_meta_L_self_modules_features_modules_6_modules_branch1_modules_0_modules_bn_parameters_bias_:
+    name = "L_self_modules_features_modules_6_modules_branch1_modules_0_modules_bn_parameters_bias_"
+    shape = [64]
+    dtype = "torch.float32"
+    device = "cpu"
+    mean = -0.080
+    std = 0.466
+    data = None
+
+
+class Program_weight_tensor_meta_L_self_modules_features_modules_6_modules_branch1_modules_1_modules_conv_parameters_weight_:
+    name = "L_self_modules_features_modules_6_modules_branch1_modules_1_modules_conv_parameters_weight_"
+    shape = [96, 64, 3, 3]
+    dtype = "torch.float32"
+    device = "cpu"
+    mean = -0.003
+    std = 0.041
+    data = None
+
+
+class Program_weight_tensor_meta_L_self_modules_features_modules_6_modules_branch1_modules_1_modules_bn_buffers_running_mean_:
+    name = "L_self_modules_features_modules_6_modules_branch1_modules_1_modules_bn_buffers_running_mean_"
+    shape = [96]
+    dtype = "torch.float32"
+    device = "cpu"
+    mean = -0.638
+    std = 0.591
+    data = None
+
+
+class Program_weight_tensor_meta_L_self_modules_features_modules_6_modules_branch1_modules_1_modules_bn_buffers_running_var_:
+    name = "L_self_modules_features_modules_6_modules_branch1_modules_1_modules_bn_buffers_running_var_"
+    shape = [96]
+    dtype = "torch.float32"
+    device = "cpu"
+    mean = 1.229
+    std = 0.802
+    data = None
+
+
+class Program_weight_tensor_meta_L_self_modules_features_modules_6_modules_branch1_modules_1_modules_bn_parameters_weight_:
+    name = "L_self_modules_features_modules_6_modules_branch1_modules_1_modules_bn_parameters_weight_"
+    shape = [96]
+    dtype = "torch.float32"
+    device = "cpu"
+    mean = 1.000
+    std = 0.000
+    data = None
+
+
+class Program_weight_tensor_meta_L_self_modules_features_modules_6_modules_branch1_modules_1_modules_bn_parameters_bias_:
+    name = "L_self_modules_features_modules_6_modules_branch1_modules_1_modules_bn_parameters_bias_"
+    shape = [96]
+    dtype = "torch.float32"
+    device = "cpu"
+    mean = -0.346
+    std = 0.501
+    data = None
+
+
+class Program_weight_tensor_meta_L_self_modules_features_modules_6_modules_branch2_modules_0_modules_conv_parameters_weight_:
+    name = "L_self_modules_features_modules_6_modules_branch2_modules_0_modules_conv_parameters_weight_"
+    shape = [64, 384, 1, 1]
+    dtype = "torch.float32"
+    device = "cpu"
+    mean = -0.002
+    std = 0.055
+    data = None
+
+
+class Program_weight_tensor_meta_L_self_modules_features_modules_6_modules_branch2_modules_0_modules_bn_buffers_running_mean_:
+    name = "L_self_modules_features_modules_6_modules_branch2_modules_0_modules_bn_buffers_running_mean_"
+    shape = [64]
+    dtype = "torch.float32"
+    device = "cpu"
+    mean = -0.102
+    std = 1.821
+    data = None
+
+
+class Program_weight_tensor_meta_L_self_modules_features_modules_6_modules_branch2_modules_0_modules_bn_buffers_running_var_:
+    name = "L_self_modules_features_modules_6_modules_branch2_modules_0_modules_bn_buffers_running_var_"
+    shape = [64]
+    dtype = "torch.float32"
+    device = "cpu"
+    mean = 1.267
+    std = 0.682
+    data = None
+
+
+class Program_weight_tensor_meta_L_self_modules_features_modules_6_modules_branch2_modules_0_modules_bn_parameters_weight_:
+    name = "L_self_modules_features_modules_6_modules_branch2_modules_0_modules_bn_parameters_weight_"
+    shape = [64]
+    dtype = "torch.float32"
+    device = "cpu"
+    mean = 1.000
+    std = 0.000
+    data = None
+
+
+class Program_weight_tensor_meta_L_self_modules_features_modules_6_modules_branch2_modules_0_modules_bn_parameters_bias_:
+    name = "L_self_modules_features_modules_6_modules_branch2_modules_0_modules_bn_parameters_bias_"
+    shape = [64]
+    dtype = "torch.float32"
+    device = "cpu"
+    mean = -0.028
+    std = 0.559
+    data = None
+
+
+class Program_weight_tensor_meta_L_self_modules_features_modules_6_modules_branch2_modules_1_modules_conv_parameters_weight_:
+    name = "L_self_modules_features_modules_6_modules_branch2_modules_1_modules_conv_parameters_weight_"
+    shape = [96, 64, 3, 3]
+    dtype = "torch.float32"
+    device = "cpu"
+    mean = -0.003
+    std = 0.044
+    data = None
+
+
+class Program_weight_tensor_meta_L_self_modules_features_modules_6_modules_branch2_modules_1_modules_bn_buffers_running_mean_:
+    name = "L_self_modules_features_modules_6_modules_branch2_modules_1_modules_bn_buffers_running_mean_"
+    shape = [96]
+    dtype = "torch.float32"
+    device = "cpu"
+    mean = -0.784
+    std = 0.422
+    data = None
+
+
+class Program_weight_tensor_meta_L_self_modules_features_modules_6_modules_branch2_modules_1_modules_bn_buffers_running_var_:
+    name = "L_self_modules_features_modules_6_modules_branch2_modules_1_modules_bn_buffers_running_var_"
+    shape = [96]
+    dtype = "torch.float32"
+    device = "cpu"
+    mean = 1.140
+    std = 0.704
+    data = None
+
+
+class Program_weight_tensor_meta_L_self_modules_features_modules_6_modules_branch2_modules_1_modules_bn_parameters_weight_:
+    name = "L_self_modules_features_modules_6_modules_branch2_modules_1_modules_bn_parameters_weight_"
+    shape = [96]
+    dtype = "torch.float32"
+    device = "cpu"
+    mean = 1.000
+    std = 0.000
+    data = None
+
+
+class Program_weight_tensor_meta_L_self_modules_features_modules_6_modules_branch2_modules_1_modules_bn_parameters_bias_:
+    name = "L_self_modules_features_modules_6_modules_branch2_modules_1_modules_bn_parameters_bias_"
+    shape = [96]
+    dtype = "torch.float32"
+    device = "cpu"
+    mean = -0.475
+    std = 0.270
+    data = None
+
+
+class Program_weight_tensor_meta_L_self_modules_features_modules_6_modules_branch2_modules_2_modules_conv_parameters_weight_:
+    name = "L_self_modules_features_modules_6_modules_branch2_modules_2_modules_conv_parameters_weight_"
+    shape = [96, 96, 3, 3]
+    dtype = "torch.float32"
+    device = "cpu"
+    mean = -0.002
+    std = 0.036
+    data = None
+
+
+class Program_weight_tensor_meta_L_self_modules_features_modules_6_modules_branch2_modules_2_modules_bn_buffers_running_mean_:
+    name = "L_self_modules_features_modules_6_modules_branch2_modules_2_modules_bn_buffers_running_mean_"
+    shape = [96]
+    dtype = "torch.float32"
+    device = "cpu"
+    mean = -0.365
+    std = 0.386
+    data = None
+
+
+class Program_weight_tensor_meta_L_self_modules_features_modules_6_modules_branch2_modules_2_modules_bn_buffers_running_var_:
+    name = "L_self_modules_features_modules_6_modules_branch2_modules_2_modules_bn_buffers_running_var_"
+    shape = [96]
+    dtype = "torch.float32"
+    device = "cpu"
+    mean = 0.716
+    std = 0.434
+    data = None
+
+
+class Program_weight_tensor_meta_L_self_modules_features_modules_6_modules_branch2_modules_2_modules_bn_parameters_weight_:
+    name = "L_self_modules_features_modules_6_modules_branch2_modules_2_modules_bn_parameters_weight_"
+    shape = [96]
+    dtype = "torch.float32"
+    device = "cpu"
+    mean = 1.000
+    std = 0.000
+    data = None
+
+
+class Program_weight_tensor_meta_L_self_modules_features_modules_6_modules_branch2_modules_2_modules_bn_parameters_bias_:
+    name = "L_self_modules_features_modules_6_modules_branch2_modules_2_modules_bn_parameters_bias_"
+    shape = [96]
+    dtype = "torch.float32"
+    device = "cpu"
+    mean = -0.245
+    std = 0.600
+    data = None
+
+
+class Program_weight_tensor_meta_L_self_modules_features_modules_6_modules_branch3_modules_1_modules_conv_parameters_weight_:
+    name = "L_self_modules_features_modules_6_modules_branch3_modules_1_modules_conv_parameters_weight_"
+    shape = [96, 384, 1, 1]
+    dtype = "torch.float32"
+    device = "cpu"
+    mean = -0.002
+    std = 0.046
+    data = None
+
+
+class Program_weight_tensor_meta_L_self_modules_features_modules_6_modules_branch3_modules_1_modules_bn_buffers_running_mean_:
+    name = "L_self_modules_features_modules_6_modules_branch3_modules_1_modules_bn_buffers_running_mean_"
+    shape = [96]
+    dtype = "torch.float32"
+    device = "cpu"
+    mean = -0.325
+    std = 1.403
+    data = None
+
+
+class Program_weight_tensor_meta_L_self_modules_features_modules_6_modules_branch3_modules_1_modules_bn_buffers_running_var_:
+    name = "L_self_modules_features_modules_6_modules_branch3_modules_1_modules_bn_buffers_running_var_"
+    shape = [96]
+    dtype = "torch.float32"
+    device = "cpu"
+    mean = 0.283
+    std = 0.129
+    data = None
+
+
+class Program_weight_tensor_meta_L_self_modules_features_modules_6_modules_branch3_modules_1_modules_bn_parameters_weight_:
+    name = "L_self_modules_features_modules_6_modules_branch3_modules_1_modules_bn_parameters_weight_"
+    shape = [96]
+    dtype = "torch.float32"
+    device = "cpu"
+    mean = 1.000
+    std = 0.000
+    data = None
+
+
+class Program_weight_tensor_meta_L_self_modules_features_modules_6_modules_branch3_modules_1_modules_bn_parameters_bias_:
+    name = "L_self_modules_features_modules_6_modules_branch3_modules_1_modules_bn_parameters_bias_"
+    shape = [96]
+    dtype = "torch.float32"
+    device = "cpu"
+    mean = -0.806
+    std = 0.351
+    data = None
+
+
+class Program_weight_tensor_meta_L_self_modules_features_modules_7_modules_branch0_modules_conv_parameters_weight_:
+    name = "L_self_modules_features_modules_7_modules_branch0_modules_conv_parameters_weight_"
+    shape = [96, 384, 1, 1]
+    dtype = "torch.float32"
+    device = "cpu"
+    mean = 0.001
+    std = 0.067
+    data = None
+
+
+class Program_weight_tensor_meta_L_self_modules_features_modules_7_modules_branch0_modules_bn_buffers_running_mean_:
+    name = "L_self_modules_features_modules_7_modules_branch0_modules_bn_buffers_running_mean_"
+    shape = [96]
+    dtype = "torch.float32"
+    device = "cpu"
+    mean = 0.250
+    std = 0.676
+    data = None
+
+
+class Program_weight_tensor_meta_L_self_modules_features_modules_7_modules_branch0_modules_bn_buffers_running_var_:
+    name = "L_self_modules_features_modules_7_modules_branch0_modules_bn_buffers_running_var_"
+    shape = [96]
+    dtype = "torch.float32"
+    device = "cpu"
+    mean = 1.243
+    std = 0.827
+    data = None
+
+
+class Program_weight_tensor_meta_L_self_modules_features_modules_7_modules_branch0_modules_bn_parameters_weight_:
+    name = "L_self_modules_features_modules_7_modules_branch0_modules_bn_parameters_weight_"
+    shape = [96]
+    dtype = "torch.float32"
+    device = "cpu"
+    mean = 1.000
+    std = 0.000
+    data = None
+
+
+class Program_weight_tensor_meta_L_self_modules_features_modules_7_modules_branch0_modules_bn_parameters_bias_:
+    name = (
+        "L_self_modules_features_modules_7_modules_branch0_modules_bn_parameters_bias_"
+    )
+    shape = [96]
+    dtype = "torch.float32"
+    device = "cpu"
+    mean = 0.316
+    std = 0.645
+    data = None
+
+
+class Program_weight_tensor_meta_L_self_modules_features_modules_7_modules_branch1_modules_0_modules_conv_parameters_weight_:
+    name = "L_self_modules_features_modules_7_modules_branch1_modules_0_modules_conv_parameters_weight_"
+    shape = [64, 384, 1, 1]
+    dtype = "torch.float32"
+    device = "cpu"
+    mean = -0.002
+    std = 0.046
+    data = None
+
+
+class Program_weight_tensor_meta_L_self_modules_features_modules_7_modules_branch1_modules_0_modules_bn_buffers_running_mean_:
+    name = "L_self_modules_features_modules_7_modules_branch1_modules_0_modules_bn_buffers_running_mean_"
+    shape = [64]
+    dtype = "torch.float32"
+    device = "cpu"
+    mean = -0.210
+    std = 0.462
+    data = None
+
+
+class Program_weight_tensor_meta_L_self_modules_features_modules_7_modules_branch1_modules_0_modules_bn_buffers_running_var_:
+    name = "L_self_modules_features_modules_7_modules_branch1_modules_0_modules_bn_buffers_running_var_"
+    shape = [64]
+    dtype = "torch.float32"
+    device = "cpu"
+    mean = 0.479
+    std = 0.223
+    data = None
+
+
+class Program_weight_tensor_meta_L_self_modules_features_modules_7_modules_branch1_modules_0_modules_bn_parameters_weight_:
+    name = "L_self_modules_features_modules_7_modules_branch1_modules_0_modules_bn_parameters_weight_"
+    shape = [64]
+    dtype = "torch.float32"
+    device = "cpu"
+    mean = 1.000
+    std = 0.000
+    data = None
+
+
+class Program_weight_tensor_meta_L_self_modules_features_modules_7_modules_branch1_modules_0_modules_bn_parameters_bias_:
+    name = "L_self_modules_features_modules_7_modules_branch1_modules_0_modules_bn_parameters_bias_"
+    shape = [64]
+    dtype = "torch.float32"
+    device = "cpu"
+    mean = -0.602
+    std = 0.375
+    data = None
+
+
+class Program_weight_tensor_meta_L_self_modules_features_modules_7_modules_branch1_modules_1_modules_conv_parameters_weight_:
+    name = "L_self_modules_features_modules_7_modules_branch1_modules_1_modules_conv_parameters_weight_"
+    shape = [96, 64, 3, 3]
+    dtype = "torch.float32"
+    device = "cpu"
+    mean = -0.003
+    std = 0.030
+    data = None
+
+
+class Program_weight_tensor_meta_L_self_modules_features_modules_7_modules_branch1_modules_1_modules_bn_buffers_running_mean_:
+    name = "L_self_modules_features_modules_7_modules_branch1_modules_1_modules_bn_buffers_running_mean_"
+    shape = [96]
+    dtype = "torch.float32"
+    device = "cpu"
+    mean = -0.320
+    std = 0.327
+    data = None
+
+
+class Program_weight_tensor_meta_L_self_modules_features_modules_7_modules_branch1_modules_1_modules_bn_buffers_running_var_:
+    name = "L_self_modules_features_modules_7_modules_branch1_modules_1_modules_bn_buffers_running_var_"
+    shape = [96]
+    dtype = "torch.float32"
+    device = "cpu"
+    mean = 0.243
+    std = 0.250
+    data = None
+
+
+class Program_weight_tensor_meta_L_self_modules_features_modules_7_modules_branch1_modules_1_modules_bn_parameters_weight_:
+    name = "L_self_modules_features_modules_7_modules_branch1_modules_1_modules_bn_parameters_weight_"
+    shape = [96]
+    dtype = "torch.float32"
+    device = "cpu"
+    mean = 1.000
+    std = 0.000
+    data = None
+
+
+class Program_weight_tensor_meta_L_self_modules_features_modules_7_modules_branch1_modules_1_modules_bn_parameters_bias_:
+    name = "L_self_modules_features_modules_7_modules_branch1_modules_1_modules_bn_parameters_bias_"
+    shape = [96]
+    dtype = "torch.float32"
+    device = "cpu"
+    mean = -0.652
+    std = 0.445
+    data = None
+
+
+class Program_weight_tensor_meta_L_self_modules_features_modules_7_modules_branch2_modules_0_modules_conv_parameters_weight_:
+    name = "L_self_modules_features_modules_7_modules_branch2_modules_0_modules_conv_parameters_weight_"
+    shape = [64, 384, 1, 1]
+    dtype = "torch.float32"
+    device = "cpu"
+    mean = -0.002
+    std = 0.049
+    data = None
+
+
+class Program_weight_tensor_meta_L_self_modules_features_modules_7_modules_branch2_modules_0_modules_bn_buffers_running_mean_:
+    name = "L_self_modules_features_modules_7_modules_branch2_modules_0_modules_bn_buffers_running_mean_"
+    shape = [64]
+    dtype = "torch.float32"
+    device = "cpu"
+    mean = -0.255
+    std = 0.675
+    data = None
+
+
+class Program_weight_tensor_meta_L_self_modules_features_modules_7_modules_branch2_modules_0_modules_bn_buffers_running_var_:
+    name = "L_self_modules_features_modules_7_modules_branch2_modules_0_modules_bn_buffers_running_var_"
+    shape = [64]
+    dtype = "torch.float32"
+    device = "cpu"
+    mean = 0.552
+    std = 0.413
+    data = None
+
+
+class Program_weight_tensor_meta_L_self_modules_features_modules_7_modules_branch2_modules_0_modules_bn_parameters_weight_:
+    name = "L_self_modules_features_modules_7_modules_branch2_modules_0_modules_bn_parameters_weight_"
+    shape = [64]
+    dtype = "torch.float32"
+    device = "cpu"
+    mean = 1.000
+    std = 0.000
+    data = None
+
+
+class Program_weight_tensor_meta_L_self_modules_features_modules_7_modules_branch2_modules_0_modules_bn_parameters_bias_:
+    name = "L_self_modules_features_modules_7_modules_branch2_modules_0_modules_bn_parameters_bias_"
+    shape = [64]
+    dtype = "torch.float32"
+    device = "cpu"
+    mean = -0.073
+    std = 0.497
+    data = None
+
+
+class Program_weight_tensor_meta_L_self_modules_features_modules_7_modules_branch2_modules_1_modules_conv_parameters_weight_:
+    name = "L_self_modules_features_modules_7_modules_branch2_modules_1_modules_conv_parameters_weight_"
+    shape = [96, 64, 3, 3]
+    dtype = "torch.float32"
+    device = "cpu"
+    mean = -0.003
+    std = 0.037
+    data = None
+
+
+class Program_weight_tensor_meta_L_self_modules_features_modules_7_modules_branch2_modules_1_modules_bn_buffers_running_mean_:
+    name = "L_self_modules_features_modules_7_modules_branch2_modules_1_modules_bn_buffers_running_mean_"
+    shape = [96]
+    dtype = "torch.float32"
+    device = "cpu"
+    mean = -0.763
+    std = 0.657
+    data = None
+
+
+class Program_weight_tensor_meta_L_self_modules_features_modules_7_modules_branch2_modules_1_modules_bn_buffers_running_var_:
+    name = "L_self_modules_features_modules_7_modules_branch2_modules_1_modules_bn_buffers_running_var_"
+    shape = [96]
+    dtype = "torch.float32"
+    device = "cpu"
+    mean = 0.666
+    std = 0.252
+    data = None
+
+
+class Program_weight_tensor_meta_L_self_modules_features_modules_7_modules_branch2_modules_1_modules_bn_parameters_weight_:
+    name = "L_self_modules_features_modules_7_modules_branch2_modules_1_modules_bn_parameters_weight_"
+    shape = [96]
+    dtype = "torch.float32"
+    device = "cpu"
+    mean = 1.000
+    std = 0.000
+    data = None
+
+
+class Program_weight_tensor_meta_L_self_modules_features_modules_7_modules_branch2_modules_1_modules_bn_parameters_bias_:
+    name = "L_self_modules_features_modules_7_modules_branch2_modules_1_modules_bn_parameters_bias_"
+    shape = [96]
+    dtype = "torch.float32"
+    device = "cpu"
+    mean = -0.609
+    std = 0.381
+    data = None
+
+
+class Program_weight_tensor_meta_L_self_modules_features_modules_7_modules_branch2_modules_2_modules_conv_parameters_weight_:
+    name = "L_self_modules_features_modules_7_modules_branch2_modules_2_modules_conv_parameters_weight_"
+    shape = [96, 96, 3, 3]
+    dtype = "torch.float32"
+    device = "cpu"
+    mean = -0.002
+    std = 0.029
+    data = None
+
+
+class Program_weight_tensor_meta_L_self_modules_features_modules_7_modules_branch2_modules_2_modules_bn_buffers_running_mean_:
+    name = "L_self_modules_features_modules_7_modules_branch2_modules_2_modules_bn_buffers_running_mean_"
+    shape = [96]
+    dtype = "torch.float32"
+    device = "cpu"
+    mean = -0.261
+    std = 0.312
+    data = None
+
+
+class Program_weight_tensor_meta_L_self_modules_features_modules_7_modules_branch2_modules_2_modules_bn_buffers_running_var_:
+    name = "L_self_modules_features_modules_7_modules_branch2_modules_2_modules_bn_buffers_running_var_"
+    shape = [96]
+    dtype = "torch.float32"
+    device = "cpu"
+    mean = 0.410
+    std = 0.320
+    data = None
+
+
+class Program_weight_tensor_meta_L_self_modules_features_modules_7_modules_branch2_modules_2_modules_bn_parameters_weight_:
+    name = "L_self_modules_features_modules_7_modules_branch2_modules_2_modules_bn_parameters_weight_"
+    shape = [96]
+    dtype = "torch.float32"
+    device = "cpu"
+    mean = 1.000
+    std = 0.000
+    data = None
+
+
+class Program_weight_tensor_meta_L_self_modules_features_modules_7_modules_branch2_modules_2_modules_bn_parameters_bias_:
+    name = "L_self_modules_features_modules_7_modules_branch2_modules_2_modules_bn_parameters_bias_"
+    shape = [96]
+    dtype = "torch.float32"
+    device = "cpu"
+    mean = -0.509
+    std = 0.566
+    data = None
+
+
+class Program_weight_tensor_meta_L_self_modules_features_modules_7_modules_branch3_modules_1_modules_conv_parameters_weight_:
+    name = "L_self_modules_features_modules_7_modules_branch3_modules_1_modules_conv_parameters_weight_"
+    shape = [96, 384, 1, 1]
+    dtype = "torch.float32"
+    device = "cpu"
+    mean = -0.002
+    std = 0.049
+    data = None
+
+
+class Program_weight_tensor_meta_L_self_modules_features_modules_7_modules_branch3_modules_1_modules_bn_buffers_running_mean_:
+    name = "L_self_modules_features_modules_7_modules_branch3_modules_1_modules_bn_buffers_running_mean_"
+    shape = [96]
+    dtype = "torch.float32"
+    device = "cpu"
+    mean = -0.297
+    std = 0.471
+    data = None
+
+
+class Program_weight_tensor_meta_L_self_modules_features_modules_7_modules_branch3_modules_1_modules_bn_buffers_running_var_:
+    name = "L_self_modules_features_modules_7_modules_branch3_modules_1_modules_bn_buffers_running_var_"
+    shape = [96]
+    dtype = "torch.float32"
+    device = "cpu"
+    mean = 0.193
+    std = 0.084
+    data = None
+
+
+class Program_weight_tensor_meta_L_self_modules_features_modules_7_modules_branch3_modules_1_modules_bn_parameters_weight_:
+    name = "L_self_modules_features_modules_7_modules_branch3_modules_1_modules_bn_parameters_weight_"
+    shape = [96]
+    dtype = "torch.float32"
+    device = "cpu"
+    mean = 1.000
+    std = 0.000
+    data = None
+
+
+class Program_weight_tensor_meta_L_self_modules_features_modules_7_modules_branch3_modules_1_modules_bn_parameters_bias_:
+    name = "L_self_modules_features_modules_7_modules_branch3_modules_1_modules_bn_parameters_bias_"
+    shape = [96]
+    dtype = "torch.float32"
+    device = "cpu"
+    mean = -0.671
+    std = 0.300
+    data = None
+
+
+class Program_weight_tensor_meta_L_self_modules_features_modules_8_modules_branch0_modules_conv_parameters_weight_:
+    name = "L_self_modules_features_modules_8_modules_branch0_modules_conv_parameters_weight_"
+    shape = [96, 384, 1, 1]
+    dtype = "torch.float32"
+    device = "cpu"
+    mean = 0.001
+    std = 0.063
+    data = None
+
+
+class Program_weight_tensor_meta_L_self_modules_features_modules_8_modules_branch0_modules_bn_buffers_running_mean_:
+    name = "L_self_modules_features_modules_8_modules_branch0_modules_bn_buffers_running_mean_"
+    shape = [96]
+    dtype = "torch.float32"
+    device = "cpu"
+    mean = 0.370
+    std = 0.676
+    data = None
+
+
+class Program_weight_tensor_meta_L_self_modules_features_modules_8_modules_branch0_modules_bn_buffers_running_var_:
+    name = "L_self_modules_features_modules_8_modules_branch0_modules_bn_buffers_running_var_"
+    shape = [96]
+    dtype = "torch.float32"
+    device = "cpu"
+    mean = 1.322
+    std = 0.829
+    data = None
+
+
+class Program_weight_tensor_meta_L_self_modules_features_modules_8_modules_branch0_modules_bn_parameters_weight_:
+    name = "L_self_modules_features_modules_8_modules_branch0_modules_bn_parameters_weight_"
+    shape = [96]
+    dtype = "torch.float32"
+    device = "cpu"
+    mean = 1.000
+    std = 0.000
+    data = None
+
+
+class Program_weight_tensor_meta_L_self_modules_features_modules_8_modules_branch0_modules_bn_parameters_bias_:
+    name = (
+        "L_self_modules_features_modules_8_modules_branch0_modules_bn_parameters_bias_"
+    )
+    shape = [96]
+    dtype = "torch.float32"
+    device = "cpu"
+    mean = 0.477
+    std = 0.723
+    data = None
+
+
+class Program_weight_tensor_meta_L_self_modules_features_modules_8_modules_branch1_modules_0_modules_conv_parameters_weight_:
+    name = "L_self_modules_features_modules_8_modules_branch1_modules_0_modules_conv_parameters_weight_"
+    shape = [64, 384, 1, 1]
+    dtype = "torch.float32"
+    device = "cpu"
+    mean = -0.003
+    std = 0.043
+    data = None
+
+
+class Program_weight_tensor_meta_L_self_modules_features_modules_8_modules_branch1_modules_0_modules_bn_buffers_running_mean_:
+    name = "L_self_modules_features_modules_8_modules_branch1_modules_0_modules_bn_buffers_running_mean_"
+    shape = [64]
+    dtype = "torch.float32"
+    device = "cpu"
+    mean = -0.154
+    std = 0.412
+    data = None
+
+
+class Program_weight_tensor_meta_L_self_modules_features_modules_8_modules_branch1_modules_0_modules_bn_buffers_running_var_:
+    name = "L_self_modules_features_modules_8_modules_branch1_modules_0_modules_bn_buffers_running_var_"
+    shape = [64]
+    dtype = "torch.float32"
+    device = "cpu"
+    mean = 0.424
+    std = 0.166
+    data = None
+
+
+class Program_weight_tensor_meta_L_self_modules_features_modules_8_modules_branch1_modules_0_modules_bn_parameters_weight_:
+    name = "L_self_modules_features_modules_8_modules_branch1_modules_0_modules_bn_parameters_weight_"
+    shape = [64]
+    dtype = "torch.float32"
+    device = "cpu"
+    mean = 1.000
+    std = 0.000
+    data = None
+
+
+class Program_weight_tensor_meta_L_self_modules_features_modules_8_modules_branch1_modules_0_modules_bn_parameters_bias_:
+    name = "L_self_modules_features_modules_8_modules_branch1_modules_0_modules_bn_parameters_bias_"
+    shape = [64]
+    dtype = "torch.float32"
+    device = "cpu"
+    mean = -0.755
+    std = 0.335
+    data = None
+
+
+class Program_weight_tensor_meta_L_self_modules_features_modules_8_modules_branch1_modules_1_modules_conv_parameters_weight_:
+    name = "L_self_modules_features_modules_8_modules_branch1_modules_1_modules_conv_parameters_weight_"
+    shape = [96, 64, 3, 3]
+    dtype = "torch.float32"
+    device = "cpu"
+    mean = -0.002
+    std = 0.027
+    data = None
+
+
+class Program_weight_tensor_meta_L_self_modules_features_modules_8_modules_branch1_modules_1_modules_bn_buffers_running_mean_:
+    name = "L_self_modules_features_modules_8_modules_branch1_modules_1_modules_bn_buffers_running_mean_"
+    shape = [96]
+    dtype = "torch.float32"
+    device = "cpu"
+    mean = -0.218
+    std = 0.232
+    data = None
+
+
+class Program_weight_tensor_meta_L_self_modules_features_modules_8_modules_branch1_modules_1_modules_bn_buffers_running_var_:
+    name = "L_self_modules_features_modules_8_modules_branch1_modules_1_modules_bn_buffers_running_var_"
+    shape = [96]
+    dtype = "torch.float32"
+    device = "cpu"
+    mean = 0.133
+    std = 0.039
+    data = None
+
+
+class Program_weight_tensor_meta_L_self_modules_features_modules_8_modules_branch1_modules_1_modules_bn_parameters_weight_:
+    name = "L_self_modules_features_modules_8_modules_branch1_modules_1_modules_bn_parameters_weight_"
+    shape = [96]
+    dtype = "torch.float32"
+    device = "cpu"
+    mean = 1.000
+    std = 0.000
+    data = None
+
+
+class Program_weight_tensor_meta_L_self_modules_features_modules_8_modules_branch1_modules_1_modules_bn_parameters_bias_:
+    name = "L_self_modules_features_modules_8_modules_branch1_modules_1_modules_bn_parameters_bias_"
+    shape = [96]
+    dtype = "torch.float32"
+    device = "cpu"
+    mean = -0.807
+    std = 0.324
+    data = None
+
+
+class Program_weight_tensor_meta_L_self_modules_features_modules_8_modules_branch2_modules_0_modules_conv_parameters_weight_:
+    name = "L_self_modules_features_modules_8_modules_branch2_modules_0_modules_conv_parameters_weight_"
+    shape = [64, 384, 1, 1]
+    dtype = "torch.float32"
+    device = "cpu"
+    mean = -0.001
+    std = 0.044
+    data = None
+
+
+class Program_weight_tensor_meta_L_self_modules_features_modules_8_modules_branch2_modules_0_modules_bn_buffers_running_mean_:
+    name = "L_self_modules_features_modules_8_modules_branch2_modules_0_modules_bn_buffers_running_mean_"
+    shape = [64]
+    dtype = "torch.float32"
+    device = "cpu"
+    mean = 0.032
+    std = 0.788
+    data = None
+
+
+class Program_weight_tensor_meta_L_self_modules_features_modules_8_modules_branch2_modules_0_modules_bn_buffers_running_var_:
+    name = "L_self_modules_features_modules_8_modules_branch2_modules_0_modules_bn_buffers_running_var_"
+    shape = [64]
+    dtype = "torch.float32"
+    device = "cpu"
+    mean = 0.502
+    std = 0.341
+    data = None
+
+
+class Program_weight_tensor_meta_L_self_modules_features_modules_8_modules_branch2_modules_0_modules_bn_parameters_weight_:
+    name = "L_self_modules_features_modules_8_modules_branch2_modules_0_modules_bn_parameters_weight_"
+    shape = [64]
+    dtype = "torch.float32"
+    device = "cpu"
+    mean = 1.000
+    std = 0.000
+    data = None
+
+
+class Program_weight_tensor_meta_L_self_modules_features_modules_8_modules_branch2_modules_0_modules_bn_parameters_bias_:
+    name = "L_self_modules_features_modules_8_modules_branch2_modules_0_modules_bn_parameters_bias_"
+    shape = [64]
+    dtype = "torch.float32"
+    device = "cpu"
+    mean = -0.416
+    std = 0.901
+    data = None
+
+
+class Program_weight_tensor_meta_L_self_modules_features_modules_8_modules_branch2_modules_1_modules_conv_parameters_weight_:
+    name = "L_self_modules_features_modules_8_modules_branch2_modules_1_modules_conv_parameters_weight_"
+    shape = [96, 64, 3, 3]
+    dtype = "torch.float32"
+    device = "cpu"
+    mean = -0.003
+    std = 0.032
+    data = None
+
+
+class Program_weight_tensor_meta_L_self_modules_features_modules_8_modules_branch2_modules_1_modules_bn_buffers_running_mean_:
+    name = "L_self_modules_features_modules_8_modules_branch2_modules_1_modules_bn_buffers_running_mean_"
+    shape = [96]
+    dtype = "torch.float32"
+    device = "cpu"
+    mean = -0.567
+    std = 1.224
+    data = None
+
+
+class Program_weight_tensor_meta_L_self_modules_features_modules_8_modules_branch2_modules_1_modules_bn_buffers_running_var_:
+    name = "L_self_modules_features_modules_8_modules_branch2_modules_1_modules_bn_buffers_running_var_"
+    shape = [96]
+    dtype = "torch.float32"
+    device = "cpu"
+    mean = 0.527
+    std = 0.413
+    data = None
+
+
+class Program_weight_tensor_meta_L_self_modules_features_modules_8_modules_branch2_modules_1_modules_bn_parameters_weight_:
+    name = "L_self_modules_features_modules_8_modules_branch2_modules_1_modules_bn_parameters_weight_"
+    shape = [96]
+    dtype = "torch.float32"
+    device = "cpu"
+    mean = 1.000
+    std = 0.000
+    data = None
+
+
+class Program_weight_tensor_meta_L_self_modules_features_modules_8_modules_branch2_modules_1_modules_bn_parameters_bias_:
+    name = "L_self_modules_features_modules_8_modules_branch2_modules_1_modules_bn_parameters_bias_"
+    shape = [96]
+    dtype = "torch.float32"
+    device = "cpu"
+    mean = -0.695
+    std = 0.655
+    data = None
+
+
+class Program_weight_tensor_meta_L_self_modules_features_modules_8_modules_branch2_modules_2_modules_conv_parameters_weight_:
+    name = "L_self_modules_features_modules_8_modules_branch2_modules_2_modules_conv_parameters_weight_"
+    shape = [96, 96, 3, 3]
+    dtype = "torch.float32"
+    device = "cpu"
+    mean = -0.001
+    std = 0.026
+    data = None
+
+
+class Program_weight_tensor_meta_L_self_modules_features_modules_8_modules_branch2_modules_2_modules_bn_buffers_running_mean_:
+    name = "L_self_modules_features_modules_8_modules_branch2_modules_2_modules_bn_buffers_running_mean_"
+    shape = [96]
+    dtype = "torch.float32"
+    device = "cpu"
+    mean = -0.277
+    std = 0.492
+    data = None
+
+
+class Program_weight_tensor_meta_L_self_modules_features_modules_8_modules_branch2_modules_2_modules_bn_buffers_running_var_:
+    name = "L_self_modules_features_modules_8_modules_branch2_modules_2_modules_bn_buffers_running_var_"
+    shape = [96]
+    dtype = "torch.float32"
+    device = "cpu"
+    mean = 0.379
+    std = 0.305
+    data = None
+
+
+class Program_weight_tensor_meta_L_self_modules_features_modules_8_modules_branch2_modules_2_modules_bn_parameters_weight_:
+    name = "L_self_modules_features_modules_8_modules_branch2_modules_2_modules_bn_parameters_weight_"
+    shape = [96]
+    dtype = "torch.float32"
+    device = "cpu"
+    mean = 1.000
+    std = 0.000
+    data = None
+
+
+class Program_weight_tensor_meta_L_self_modules_features_modules_8_modules_branch2_modules_2_modules_bn_parameters_bias_:
+    name = "L_self_modules_features_modules_8_modules_branch2_modules_2_modules_bn_parameters_bias_"
+    shape = [96]
+    dtype = "torch.float32"
+    device = "cpu"
+    mean = -0.571
+    std = 0.707
+    data = None
+
+
+class Program_weight_tensor_meta_L_self_modules_features_modules_8_modules_branch3_modules_1_modules_conv_parameters_weight_:
+    name = "L_self_modules_features_modules_8_modules_branch3_modules_1_modules_conv_parameters_weight_"
+    shape = [96, 384, 1, 1]
+    dtype = "torch.float32"
+    device = "cpu"
+    mean = -0.002
+    std = 0.045
+    data = None
+
+
+class Program_weight_tensor_meta_L_self_modules_features_modules_8_modules_branch3_modules_1_modules_bn_buffers_running_mean_:
+    name = "L_self_modules_features_modules_8_modules_branch3_modules_1_modules_bn_buffers_running_mean_"
+    shape = [96]
+    dtype = "torch.float32"
+    device = "cpu"
+    mean = -0.132
+    std = 0.354
+    data = None
+
+
+class Program_weight_tensor_meta_L_self_modules_features_modules_8_modules_branch3_modules_1_modules_bn_buffers_running_var_:
+    name = "L_self_modules_features_modules_8_modules_branch3_modules_1_modules_bn_buffers_running_var_"
+    shape = [96]
+    dtype = "torch.float32"
+    device = "cpu"
+    mean = 0.198
+    std = 0.102
+    data = None
+
+
+class Program_weight_tensor_meta_L_self_modules_features_modules_8_modules_branch3_modules_1_modules_bn_parameters_weight_:
+    name = "L_self_modules_features_modules_8_modules_branch3_modules_1_modules_bn_parameters_weight_"
+    shape = [96]
+    dtype = "torch.float32"
+    device = "cpu"
+    mean = 1.000
+    std = 0.000
+    data = None
+
+
+class Program_weight_tensor_meta_L_self_modules_features_modules_8_modules_branch3_modules_1_modules_bn_parameters_bias_:
+    name = "L_self_modules_features_modules_8_modules_branch3_modules_1_modules_bn_parameters_bias_"
+    shape = [96]
+    dtype = "torch.float32"
+    device = "cpu"
+    mean = -0.575
+    std = 0.293
+    data = None
+
+
+class Program_weight_tensor_meta_L_self_modules_features_modules_9_modules_branch0_modules_conv_parameters_weight_:
+    name = "L_self_modules_features_modules_9_modules_branch0_modules_conv_parameters_weight_"
+    shape = [96, 384, 1, 1]
+    dtype = "torch.float32"
+    device = "cpu"
+    mean = -0.002
+    std = 0.055
+    data = None
+
+
+class Program_weight_tensor_meta_L_self_modules_features_modules_9_modules_branch0_modules_bn_buffers_running_mean_:
+    name = "L_self_modules_features_modules_9_modules_branch0_modules_bn_buffers_running_mean_"
+    shape = [96]
+    dtype = "torch.float32"
+    device = "cpu"
+    mean = 0.063
+    std = 0.813
+    data = None
+
+
+class Program_weight_tensor_meta_L_self_modules_features_modules_9_modules_branch0_modules_bn_buffers_running_var_:
+    name = "L_self_modules_features_modules_9_modules_branch0_modules_bn_buffers_running_var_"
+    shape = [96]
+    dtype = "torch.float32"
+    device = "cpu"
+    mean = 1.054
+    std = 0.353
+    data = None
+
+
+class Program_weight_tensor_meta_L_self_modules_features_modules_9_modules_branch0_modules_bn_parameters_weight_:
+    name = "L_self_modules_features_modules_9_modules_branch0_modules_bn_parameters_weight_"
+    shape = [96]
+    dtype = "torch.float32"
+    device = "cpu"
+    mean = 1.000
+    std = 0.000
+    data = None
+
+
+class Program_weight_tensor_meta_L_self_modules_features_modules_9_modules_branch0_modules_bn_parameters_bias_:
+    name = (
+        "L_self_modules_features_modules_9_modules_branch0_modules_bn_parameters_bias_"
+    )
+    shape = [96]
+    dtype = "torch.float32"
+    device = "cpu"
+    mean = -0.102
+    std = 0.355
+    data = None
+
+
+class Program_weight_tensor_meta_L_self_modules_features_modules_9_modules_branch1_modules_0_modules_conv_parameters_weight_:
+    name = "L_self_modules_features_modules_9_modules_branch1_modules_0_modules_conv_parameters_weight_"
+    shape = [64, 384, 1, 1]
+    dtype = "torch.float32"
+    device = "cpu"
+    mean = -0.002
+    std = 0.047
+    data = None
+
+
+class Program_weight_tensor_meta_L_self_modules_features_modules_9_modules_branch1_modules_0_modules_bn_buffers_running_mean_:
+    name = "L_self_modules_features_modules_9_modules_branch1_modules_0_modules_bn_buffers_running_mean_"
+    shape = [64]
+    dtype = "torch.float32"
+    device = "cpu"
+    mean = -0.098
+    std = 0.410
+    data = None
+
+
+class Program_weight_tensor_meta_L_self_modules_features_modules_9_modules_branch1_modules_0_modules_bn_buffers_running_var_:
+    name = "L_self_modules_features_modules_9_modules_branch1_modules_0_modules_bn_buffers_running_var_"
+    shape = [64]
+    dtype = "torch.float32"
+    device = "cpu"
+    mean = 0.660
+    std = 0.246
+    data = None
+
+
+class Program_weight_tensor_meta_L_self_modules_features_modules_9_modules_branch1_modules_0_modules_bn_parameters_weight_:
+    name = "L_self_modules_features_modules_9_modules_branch1_modules_0_modules_bn_parameters_weight_"
+    shape = [64]
+    dtype = "torch.float32"
+    device = "cpu"
+    mean = 1.000
+    std = 0.000
+    data = None
+
+
+class Program_weight_tensor_meta_L_self_modules_features_modules_9_modules_branch1_modules_0_modules_bn_parameters_bias_:
+    name = "L_self_modules_features_modules_9_modules_branch1_modules_0_modules_bn_parameters_bias_"
+    shape = [64]
+    dtype = "torch.float32"
+    device = "cpu"
+    mean = -0.547
+    std = 0.357
+    data = None
+
+
+class Program_weight_tensor_meta_L_self_modules_features_modules_9_modules_branch1_modules_1_modules_conv_parameters_weight_:
+    name = "L_self_modules_features_modules_9_modules_branch1_modules_1_modules_conv_parameters_weight_"
+    shape = [96, 64, 3, 3]
+    dtype = "torch.float32"
+    device = "cpu"
+    mean = -0.003
+    std = 0.032
+    data = None
+
+
+class Program_weight_tensor_meta_L_self_modules_features_modules_9_modules_branch1_modules_1_modules_bn_buffers_running_mean_:
+    name = "L_self_modules_features_modules_9_modules_branch1_modules_1_modules_bn_buffers_running_mean_"
+    shape = [96]
+    dtype = "torch.float32"
+    device = "cpu"
+    mean = -0.347
+    std = 0.254
+    data = None
+
+
+class Program_weight_tensor_meta_L_self_modules_features_modules_9_modules_branch1_modules_1_modules_bn_buffers_running_var_:
+    name = "L_self_modules_features_modules_9_modules_branch1_modules_1_modules_bn_buffers_running_var_"
+    shape = [96]
+    dtype = "torch.float32"
+    device = "cpu"
+    mean = 0.294
+    std = 0.078
+    data = None
+
+
+class Program_weight_tensor_meta_L_self_modules_features_modules_9_modules_branch1_modules_1_modules_bn_parameters_weight_:
+    name = "L_self_modules_features_modules_9_modules_branch1_modules_1_modules_bn_parameters_weight_"
+    shape = [96]
+    dtype = "torch.float32"
+    device = "cpu"
+    mean = 1.000
+    std = 0.000
+    data = None
+
+
+class Program_weight_tensor_meta_L_self_modules_features_modules_9_modules_branch1_modules_1_modules_bn_parameters_bias_:
+    name = "L_self_modules_features_modules_9_modules_branch1_modules_1_modules_bn_parameters_bias_"
+    shape = [96]
+    dtype = "torch.float32"
+    device = "cpu"
+    mean = -0.787
+    std = 0.520
+    data = None
+
+
+class Program_weight_tensor_meta_L_self_modules_features_modules_9_modules_branch2_modules_0_modules_conv_parameters_weight_:
+    name = "L_self_modules_features_modules_9_modules_branch2_modules_0_modules_conv_parameters_weight_"
+    shape = [64, 384, 1, 1]
+    dtype = "torch.float32"
+    device = "cpu"
+    mean = -0.002
+    std = 0.041
+    data = None
+
+
+class Program_weight_tensor_meta_L_self_modules_features_modules_9_modules_branch2_modules_0_modules_bn_buffers_running_mean_:
+    name = "L_self_modules_features_modules_9_modules_branch2_modules_0_modules_bn_buffers_running_mean_"
+    shape = [64]
+    dtype = "torch.float32"
+    device = "cpu"
+    mean = -0.126
+    std = 0.786
+    data = None
+
+
+class Program_weight_tensor_meta_L_self_modules_features_modules_9_modules_branch2_modules_0_modules_bn_buffers_running_var_:
+    name = "L_self_modules_features_modules_9_modules_branch2_modules_0_modules_bn_buffers_running_var_"
+    shape = [64]
+    dtype = "torch.float32"
+    device = "cpu"
+    mean = 0.572
+    std = 0.400
+    data = None
+
+
+class Program_weight_tensor_meta_L_self_modules_features_modules_9_modules_branch2_modules_0_modules_bn_parameters_weight_:
+    name = "L_self_modules_features_modules_9_modules_branch2_modules_0_modules_bn_parameters_weight_"
+    shape = [64]
+    dtype = "torch.float32"
+    device = "cpu"
+    mean = 1.000
+    std = 0.000
+    data = None
+
+
+class Program_weight_tensor_meta_L_self_modules_features_modules_9_modules_branch2_modules_0_modules_bn_parameters_bias_:
+    name = "L_self_modules_features_modules_9_modules_branch2_modules_0_modules_bn_parameters_bias_"
+    shape = [64]
+    dtype = "torch.float32"
+    device = "cpu"
+    mean = -0.359
+    std = 0.666
+    data = None
+
+
+class Program_weight_tensor_meta_L_self_modules_features_modules_9_modules_branch2_modules_1_modules_conv_parameters_weight_:
+    name = "L_self_modules_features_modules_9_modules_branch2_modules_1_modules_conv_parameters_weight_"
+    shape = [96, 64, 3, 3]
+    dtype = "torch.float32"
+    device = "cpu"
+    mean = -0.003
+    std = 0.037
+    data = None
+
+
+class Program_weight_tensor_meta_L_self_modules_features_modules_9_modules_branch2_modules_1_modules_bn_buffers_running_mean_:
+    name = "L_self_modules_features_modules_9_modules_branch2_modules_1_modules_bn_buffers_running_mean_"
+    shape = [96]
+    dtype = "torch.float32"
+    device = "cpu"
+    mean = -0.676
+    std = 0.306
+    data = None
+
+
+class Program_weight_tensor_meta_L_self_modules_features_modules_9_modules_branch2_modules_1_modules_bn_buffers_running_var_:
+    name = "L_self_modules_features_modules_9_modules_branch2_modules_1_modules_bn_buffers_running_var_"
+    shape = [96]
+    dtype = "torch.float32"
+    device = "cpu"
+    mean = 0.888
+    std = 0.301
+    data = None
+
+
+class Program_weight_tensor_meta_L_self_modules_features_modules_9_modules_branch2_modules_1_modules_bn_parameters_weight_:
+    name = "L_self_modules_features_modules_9_modules_branch2_modules_1_modules_bn_parameters_weight_"
+    shape = [96]
+    dtype = "torch.float32"
+    device = "cpu"
+    mean = 1.000
+    std = 0.000
+    data = None
+
+
+class Program_weight_tensor_meta_L_self_modules_features_modules_9_modules_branch2_modules_1_modules_bn_parameters_bias_:
+    name = "L_self_modules_features_modules_9_modules_branch2_modules_1_modules_bn_parameters_bias_"
+    shape = [96]
+    dtype = "torch.float32"
+    device = "cpu"
+    mean = -0.723
+    std = 0.370
+    data = None
+
+
+class Program_weight_tensor_meta_L_self_modules_features_modules_9_modules_branch2_modules_2_modules_conv_parameters_weight_:
+    name = "L_self_modules_features_modules_9_modules_branch2_modules_2_modules_conv_parameters_weight_"
+    shape = [96, 96, 3, 3]
+    dtype = "torch.float32"
+    device = "cpu"
+    mean = -0.003
+    std = 0.031
+    data = None
+
+
+class Program_weight_tensor_meta_L_self_modules_features_modules_9_modules_branch2_modules_2_modules_bn_buffers_running_mean_:
+    name = "L_self_modules_features_modules_9_modules_branch2_modules_2_modules_bn_buffers_running_mean_"
+    shape = [96]
+    dtype = "torch.float32"
+    device = "cpu"
+    mean = -0.388
+    std = 0.203
+    data = None
+
+
+class Program_weight_tensor_meta_L_self_modules_features_modules_9_modules_branch2_modules_2_modules_bn_buffers_running_var_:
+    name = "L_self_modules_features_modules_9_modules_branch2_modules_2_modules_bn_buffers_running_var_"
+    shape = [96]
+    dtype = "torch.float32"
+    device = "cpu"
+    mean = 0.390
+    std = 0.206
+    data = None
+
+
+class Program_weight_tensor_meta_L_self_modules_features_modules_9_modules_branch2_modules_2_modules_bn_parameters_weight_:
+    name = "L_self_modules_features_modules_9_modules_branch2_modules_2_modules_bn_parameters_weight_"
+    shape = [96]
+    dtype = "torch.float32"
+    device = "cpu"
+    mean = 1.000
+    std = 0.000
+    data = None
+
+
+class Program_weight_tensor_meta_L_self_modules_features_modules_9_modules_branch2_modules_2_modules_bn_parameters_bias_:
+    name = "L_self_modules_features_modules_9_modules_branch2_modules_2_modules_bn_parameters_bias_"
+    shape = [96]
+    dtype = "torch.float32"
+    device = "cpu"
+    mean = -0.474
+    std = 0.936
+    data = None
+
+
+class Program_weight_tensor_meta_L_self_modules_features_modules_9_modules_branch3_modules_1_modules_conv_parameters_weight_:
+    name = "L_self_modules_features_modules_9_modules_branch3_modules_1_modules_conv_parameters_weight_"
+    shape = [96, 384, 1, 1]
+    dtype = "torch.float32"
+    device = "cpu"
+    mean = -0.001
+    std = 0.053
+    data = None
+
+
+class Program_weight_tensor_meta_L_self_modules_features_modules_9_modules_branch3_modules_1_modules_bn_buffers_running_mean_:
+    name = "L_self_modules_features_modules_9_modules_branch3_modules_1_modules_bn_buffers_running_mean_"
+    shape = [96]
+    dtype = "torch.float32"
+    device = "cpu"
+    mean = 0.118
+    std = 0.754
+    data = None
+
+
+class Program_weight_tensor_meta_L_self_modules_features_modules_9_modules_branch3_modules_1_modules_bn_buffers_running_var_:
+    name = "L_self_modules_features_modules_9_modules_branch3_modules_1_modules_bn_buffers_running_var_"
+    shape = [96]
+    dtype = "torch.float32"
+    device = "cpu"
+    mean = 0.394
+    std = 0.274
+    data = None
+
+
+class Program_weight_tensor_meta_L_self_modules_features_modules_9_modules_branch3_modules_1_modules_bn_parameters_weight_:
+    name = "L_self_modules_features_modules_9_modules_branch3_modules_1_modules_bn_parameters_weight_"
+    shape = [96]
+    dtype = "torch.float32"
+    device = "cpu"
+    mean = 1.000
+    std = 0.000
+    data = None
+
+
+class Program_weight_tensor_meta_L_self_modules_features_modules_9_modules_branch3_modules_1_modules_bn_parameters_bias_:
+    name = "L_self_modules_features_modules_9_modules_branch3_modules_1_modules_bn_parameters_bias_"
+    shape = [96]
+    dtype = "torch.float32"
+    device = "cpu"
+    mean = -0.236
+    std = 0.401
+    data = None
+
+
+class Program_weight_tensor_meta_L_self_modules_features_modules_10_modules_branch0_modules_conv_parameters_weight_:
+    name = "L_self_modules_features_modules_10_modules_branch0_modules_conv_parameters_weight_"
+    shape = [384, 384, 3, 3]
+    dtype = "torch.float32"
+    device = "cpu"
+    mean = -0.000
+    std = 0.011
+    data = None
+
+
+class Program_weight_tensor_meta_L_self_modules_features_modules_10_modules_branch0_modules_bn_buffers_running_mean_:
+    name = "L_self_modules_features_modules_10_modules_branch0_modules_bn_buffers_running_mean_"
+    shape = [384]
+    dtype = "torch.float32"
+    device = "cpu"
+    mean = -0.266
+    std = 0.784
+    data = None
+
+
+class Program_weight_tensor_meta_L_self_modules_features_modules_10_modules_branch0_modules_bn_buffers_running_var_:
+    name = "L_self_modules_features_modules_10_modules_branch0_modules_bn_buffers_running_var_"
+    shape = [384]
+    dtype = "torch.float32"
+    device = "cpu"
+    mean = 0.716
+    std = 0.215
+    data = None
+
+
+class Program_weight_tensor_meta_L_self_modules_features_modules_10_modules_branch0_modules_bn_parameters_weight_:
+    name = "L_self_modules_features_modules_10_modules_branch0_modules_bn_parameters_weight_"
+    shape = [384]
+    dtype = "torch.float32"
+    device = "cpu"
+    mean = 1.000
+    std = 0.000
+    data = None
+
+
+class Program_weight_tensor_meta_L_self_modules_features_modules_10_modules_branch0_modules_bn_parameters_bias_:
+    name = (
+        "L_self_modules_features_modules_10_modules_branch0_modules_bn_parameters_bias_"
+    )
+    shape = [384]
+    dtype = "torch.float32"
+    device = "cpu"
+    mean = -0.744
+    std = 0.226
+    data = None
+
+
+class Program_weight_tensor_meta_L_self_modules_features_modules_10_modules_branch1_modules_0_modules_conv_parameters_weight_:
+    name = "L_self_modules_features_modules_10_modules_branch1_modules_0_modules_conv_parameters_weight_"
+    shape = [192, 384, 1, 1]
+    dtype = "torch.float32"
+    device = "cpu"
+    mean = -0.001
+    std = 0.026
+    data = None
+
+
+class Program_weight_tensor_meta_L_self_modules_features_modules_10_modules_branch1_modules_0_modules_bn_buffers_running_mean_:
+    name = "L_self_modules_features_modules_10_modules_branch1_modules_0_modules_bn_buffers_running_mean_"
+    shape = [192]
+    dtype = "torch.float32"
+    device = "cpu"
+    mean = -0.052
+    std = 0.363
+    data = None
+
+
+class Program_weight_tensor_meta_L_self_modules_features_modules_10_modules_branch1_modules_0_modules_bn_buffers_running_var_:
+    name = "L_self_modules_features_modules_10_modules_branch1_modules_0_modules_bn_buffers_running_var_"
+    shape = [192]
+    dtype = "torch.float32"
+    device = "cpu"
+    mean = 0.213
+    std = 0.109
+    data = None
+
+
+class Program_weight_tensor_meta_L_self_modules_features_modules_10_modules_branch1_modules_0_modules_bn_parameters_weight_:
+    name = "L_self_modules_features_modules_10_modules_branch1_modules_0_modules_bn_parameters_weight_"
+    shape = [192]
+    dtype = "torch.float32"
+    device = "cpu"
+    mean = 1.000
+    std = 0.000
+    data = None
+
+
+class Program_weight_tensor_meta_L_self_modules_features_modules_10_modules_branch1_modules_0_modules_bn_parameters_bias_:
+    name = "L_self_modules_features_modules_10_modules_branch1_modules_0_modules_bn_parameters_bias_"
+    shape = [192]
+    dtype = "torch.float32"
+    device = "cpu"
+    mean = -0.608
+    std = 0.368
+    data = None
+
+
+class Program_weight_tensor_meta_L_self_modules_features_modules_10_modules_branch1_modules_1_modules_conv_parameters_weight_:
+    name = "L_self_modules_features_modules_10_modules_branch1_modules_1_modules_conv_parameters_weight_"
+    shape = [224, 192, 3, 3]
+    dtype = "torch.float32"
+    device = "cpu"
+    mean = -0.001
+    std = 0.014
+    data = None
+
+
+class Program_weight_tensor_meta_L_self_modules_features_modules_10_modules_branch1_modules_1_modules_bn_buffers_running_mean_:
+    name = "L_self_modules_features_modules_10_modules_branch1_modules_1_modules_bn_buffers_running_mean_"
+    shape = [224]
+    dtype = "torch.float32"
+    device = "cpu"
+    mean = -0.416
+    std = 0.309
+    data = None
+
+
+class Program_weight_tensor_meta_L_self_modules_features_modules_10_modules_branch1_modules_1_modules_bn_buffers_running_var_:
+    name = "L_self_modules_features_modules_10_modules_branch1_modules_1_modules_bn_buffers_running_var_"
+    shape = [224]
+    dtype = "torch.float32"
+    device = "cpu"
+    mean = 0.394
+    std = 0.130
+    data = None
+
+
+class Program_weight_tensor_meta_L_self_modules_features_modules_10_modules_branch1_modules_1_modules_bn_parameters_weight_:
+    name = "L_self_modules_features_modules_10_modules_branch1_modules_1_modules_bn_parameters_weight_"
+    shape = [224]
+    dtype = "torch.float32"
+    device = "cpu"
+    mean = 1.000
+    std = 0.000
+    data = None
+
+
+class Program_weight_tensor_meta_L_self_modules_features_modules_10_modules_branch1_modules_1_modules_bn_parameters_bias_:
+    name = "L_self_modules_features_modules_10_modules_branch1_modules_1_modules_bn_parameters_bias_"
+    shape = [224]
+    dtype = "torch.float32"
+    device = "cpu"
+    mean = -0.860
+    std = 0.214
+    data = None
+
+
+class Program_weight_tensor_meta_L_self_modules_features_modules_10_modules_branch1_modules_2_modules_conv_parameters_weight_:
+    name = "L_self_modules_features_modules_10_modules_branch1_modules_2_modules_conv_parameters_weight_"
+    shape = [256, 224, 3, 3]
+    dtype = "torch.float32"
+    device = "cpu"
+    mean = -0.001
+    std = 0.012
+    data = None
+
+
+class Program_weight_tensor_meta_L_self_modules_features_modules_10_modules_branch1_modules_2_modules_bn_buffers_running_mean_:
+    name = "L_self_modules_features_modules_10_modules_branch1_modules_2_modules_bn_buffers_running_mean_"
+    shape = [256]
+    dtype = "torch.float32"
+    device = "cpu"
+    mean = -0.195
+    std = 0.197
+    data = None
+
+
+class Program_weight_tensor_meta_L_self_modules_features_modules_10_modules_branch1_modules_2_modules_bn_buffers_running_var_:
+    name = "L_self_modules_features_modules_10_modules_branch1_modules_2_modules_bn_buffers_running_var_"
+    shape = [256]
+    dtype = "torch.float32"
+    device = "cpu"
+    mean = 0.188
+    std = 0.127
+    data = None
+
+
+class Program_weight_tensor_meta_L_self_modules_features_modules_10_modules_branch1_modules_2_modules_bn_parameters_weight_:
+    name = "L_self_modules_features_modules_10_modules_branch1_modules_2_modules_bn_parameters_weight_"
+    shape = [256]
+    dtype = "torch.float32"
+    device = "cpu"
+    mean = 1.000
+    std = 0.000
+    data = None
+
+
+class Program_weight_tensor_meta_L_self_modules_features_modules_10_modules_branch1_modules_2_modules_bn_parameters_bias_:
+    name = "L_self_modules_features_modules_10_modules_branch1_modules_2_modules_bn_parameters_bias_"
+    shape = [256]
+    dtype = "torch.float32"
+    device = "cpu"
+    mean = -0.333
+    std = 0.409
+    data = None
+
+
+class Program_weight_tensor_meta_L_self_modules_features_modules_11_modules_branch0_modules_conv_parameters_weight_:
+    name = "L_self_modules_features_modules_11_modules_branch0_modules_conv_parameters_weight_"
+    shape = [384, 1024, 1, 1]
+    dtype = "torch.float32"
+    device = "cpu"
+    mean = -0.001
+    std = 0.028
+    data = None
+
+
+class Program_weight_tensor_meta_L_self_modules_features_modules_11_modules_branch0_modules_bn_buffers_running_mean_:
+    name = "L_self_modules_features_modules_11_modules_branch0_modules_bn_buffers_running_mean_"
+    shape = [384]
+    dtype = "torch.float32"
+    device = "cpu"
+    mean = -0.176
+    std = 0.711
+    data = None
+
+
+class Program_weight_tensor_meta_L_self_modules_features_modules_11_modules_branch0_modules_bn_buffers_running_var_:
+    name = "L_self_modules_features_modules_11_modules_branch0_modules_bn_buffers_running_var_"
+    shape = [384]
+    dtype = "torch.float32"
+    device = "cpu"
+    mean = 0.991
+    std = 1.109
+    data = None
+
+
+class Program_weight_tensor_meta_L_self_modules_features_modules_11_modules_branch0_modules_bn_parameters_weight_:
+    name = "L_self_modules_features_modules_11_modules_branch0_modules_bn_parameters_weight_"
+    shape = [384]
+    dtype = "torch.float32"
+    device = "cpu"
+    mean = 1.000
+    std = 0.000
+    data = None
+
+
+class Program_weight_tensor_meta_L_self_modules_features_modules_11_modules_branch0_modules_bn_parameters_bias_:
+    name = (
+        "L_self_modules_features_modules_11_modules_branch0_modules_bn_parameters_bias_"
+    )
+    shape = [384]
+    dtype = "torch.float32"
+    device = "cpu"
+    mean = -0.292
+    std = 0.449
+    data = None
+
+
+class Program_weight_tensor_meta_L_self_modules_features_modules_11_modules_branch1_modules_0_modules_conv_parameters_weight_:
+    name = "L_self_modules_features_modules_11_modules_branch1_modules_0_modules_conv_parameters_weight_"
+    shape = [192, 1024, 1, 1]
+    dtype = "torch.float32"
+    device = "cpu"
+    mean = -0.001
+    std = 0.025
+    data = None
+
+
+class Program_weight_tensor_meta_L_self_modules_features_modules_11_modules_branch1_modules_0_modules_bn_buffers_running_mean_:
+    name = "L_self_modules_features_modules_11_modules_branch1_modules_0_modules_bn_buffers_running_mean_"
+    shape = [192]
+    dtype = "torch.float32"
+    device = "cpu"
+    mean = -0.069
+    std = 0.572
+    data = None
+
+
+class Program_weight_tensor_meta_L_self_modules_features_modules_11_modules_branch1_modules_0_modules_bn_buffers_running_var_:
+    name = "L_self_modules_features_modules_11_modules_branch1_modules_0_modules_bn_buffers_running_var_"
+    shape = [192]
+    dtype = "torch.float32"
+    device = "cpu"
+    mean = 0.667
+    std = 0.321
+    data = None
+
+
+class Program_weight_tensor_meta_L_self_modules_features_modules_11_modules_branch1_modules_0_modules_bn_parameters_weight_:
+    name = "L_self_modules_features_modules_11_modules_branch1_modules_0_modules_bn_parameters_weight_"
+    shape = [192]
+    dtype = "torch.float32"
+    device = "cpu"
+    mean = 1.000
+    std = 0.000
+    data = None
+
+
+class Program_weight_tensor_meta_L_self_modules_features_modules_11_modules_branch1_modules_0_modules_bn_parameters_bias_:
+    name = "L_self_modules_features_modules_11_modules_branch1_modules_0_modules_bn_parameters_bias_"
+    shape = [192]
+    dtype = "torch.float32"
+    device = "cpu"
+    mean = -0.761
+    std = 0.280
+    data = None
+
+
+class Program_weight_tensor_meta_L_self_modules_features_modules_11_modules_branch1_modules_1_modules_conv_parameters_weight_:
+    name = "L_self_modules_features_modules_11_modules_branch1_modules_1_modules_conv_parameters_weight_"
+    shape = [224, 192, 1, 7]
+    dtype = "torch.float32"
+    device = "cpu"
+    mean = -0.001
+    std = 0.022
+    data = None
+
+
+class Program_weight_tensor_meta_L_self_modules_features_modules_11_modules_branch1_modules_1_modules_bn_buffers_running_mean_:
+    name = "L_self_modules_features_modules_11_modules_branch1_modules_1_modules_bn_buffers_running_mean_"
+    shape = [224]
+    dtype = "torch.float32"
+    device = "cpu"
+    mean = -0.264
+    std = 0.190
+    data = None
+
+
+class Program_weight_tensor_meta_L_self_modules_features_modules_11_modules_branch1_modules_1_modules_bn_buffers_running_var_:
+    name = "L_self_modules_features_modules_11_modules_branch1_modules_1_modules_bn_buffers_running_var_"
+    shape = [224]
+    dtype = "torch.float32"
+    device = "cpu"
+    mean = 0.304
+    std = 0.135
+    data = None
+
+
+class Program_weight_tensor_meta_L_self_modules_features_modules_11_modules_branch1_modules_1_modules_bn_parameters_weight_:
+    name = "L_self_modules_features_modules_11_modules_branch1_modules_1_modules_bn_parameters_weight_"
+    shape = [224]
+    dtype = "torch.float32"
+    device = "cpu"
+    mean = 1.000
+    std = 0.000
+    data = None
+
+
+class Program_weight_tensor_meta_L_self_modules_features_modules_11_modules_branch1_modules_1_modules_bn_parameters_bias_:
+    name = "L_self_modules_features_modules_11_modules_branch1_modules_1_modules_bn_parameters_bias_"
+    shape = [224]
+    dtype = "torch.float32"
+    device = "cpu"
+    mean = -0.576
+    std = 0.282
+    data = None
+
+
+class Program_weight_tensor_meta_L_self_modules_features_modules_11_modules_branch1_modules_2_modules_conv_parameters_weight_:
+    name = "L_self_modules_features_modules_11_modules_branch1_modules_2_modules_conv_parameters_weight_"
+    shape = [256, 224, 7, 1]
+    dtype = "torch.float32"
+    device = "cpu"
+    mean = -0.001
+    std = 0.021
+    data = None
+
+
+class Program_weight_tensor_meta_L_self_modules_features_modules_11_modules_branch1_modules_2_modules_bn_buffers_running_mean_:
+    name = "L_self_modules_features_modules_11_modules_branch1_modules_2_modules_bn_buffers_running_mean_"
+    shape = [256]
+    dtype = "torch.float32"
+    device = "cpu"
+    mean = -0.365
+    std = 0.285
+    data = None
+
+
+class Program_weight_tensor_meta_L_self_modules_features_modules_11_modules_branch1_modules_2_modules_bn_buffers_running_var_:
+    name = "L_self_modules_features_modules_11_modules_branch1_modules_2_modules_bn_buffers_running_var_"
+    shape = [256]
+    dtype = "torch.float32"
+    device = "cpu"
+    mean = 0.331
+    std = 0.128
+    data = None
+
+
+class Program_weight_tensor_meta_L_self_modules_features_modules_11_modules_branch1_modules_2_modules_bn_parameters_weight_:
+    name = "L_self_modules_features_modules_11_modules_branch1_modules_2_modules_bn_parameters_weight_"
+    shape = [256]
+    dtype = "torch.float32"
+    device = "cpu"
+    mean = 1.000
+    std = 0.000
+    data = None
+
+
+class Program_weight_tensor_meta_L_self_modules_features_modules_11_modules_branch1_modules_2_modules_bn_parameters_bias_:
+    name = "L_self_modules_features_modules_11_modules_branch1_modules_2_modules_bn_parameters_bias_"
+    shape = [256]
+    dtype = "torch.float32"
+    device = "cpu"
+    mean = -0.650
+    std = 0.314
+    data = None
+
+
+class Program_weight_tensor_meta_L_self_modules_features_modules_11_modules_branch2_modules_0_modules_conv_parameters_weight_:
+    name = "L_self_modules_features_modules_11_modules_branch2_modules_0_modules_conv_parameters_weight_"
+    shape = [192, 1024, 1, 1]
+    dtype = "torch.float32"
+    device = "cpu"
+    mean = -0.001
+    std = 0.022
+    data = None
+
+
+class Program_weight_tensor_meta_L_self_modules_features_modules_11_modules_branch2_modules_0_modules_bn_buffers_running_mean_:
+    name = "L_self_modules_features_modules_11_modules_branch2_modules_0_modules_bn_buffers_running_mean_"
+    shape = [192]
+    dtype = "torch.float32"
+    device = "cpu"
+    mean = 0.048
+    std = 0.534
+    data = None
+
+
+class Program_weight_tensor_meta_L_self_modules_features_modules_11_modules_branch2_modules_0_modules_bn_buffers_running_var_:
+    name = "L_self_modules_features_modules_11_modules_branch2_modules_0_modules_bn_buffers_running_var_"
+    shape = [192]
+    dtype = "torch.float32"
+    device = "cpu"
+    mean = 0.669
+    std = 0.630
+    data = None
+
+
+class Program_weight_tensor_meta_L_self_modules_features_modules_11_modules_branch2_modules_0_modules_bn_parameters_weight_:
+    name = "L_self_modules_features_modules_11_modules_branch2_modules_0_modules_bn_parameters_weight_"
+    shape = [192]
+    dtype = "torch.float32"
+    device = "cpu"
+    mean = 1.000
+    std = 0.000
+    data = None
+
+
+class Program_weight_tensor_meta_L_self_modules_features_modules_11_modules_branch2_modules_0_modules_bn_parameters_bias_:
+    name = "L_self_modules_features_modules_11_modules_branch2_modules_0_modules_bn_parameters_bias_"
+    shape = [192]
+    dtype = "torch.float32"
+    device = "cpu"
+    mean = -0.785
+    std = 0.310
+    data = None
+
+
+class Program_weight_tensor_meta_L_self_modules_features_modules_11_modules_branch2_modules_1_modules_conv_parameters_weight_:
+    name = "L_self_modules_features_modules_11_modules_branch2_modules_1_modules_conv_parameters_weight_"
+    shape = [192, 192, 7, 1]
+    dtype = "torch.float32"
+    device = "cpu"
+    mean = -0.002
+    std = 0.020
+    data = None
+
+
+class Program_weight_tensor_meta_L_self_modules_features_modules_11_modules_branch2_modules_1_modules_bn_buffers_running_mean_:
+    name = "L_self_modules_features_modules_11_modules_branch2_modules_1_modules_bn_buffers_running_mean_"
+    shape = [192]
+    dtype = "torch.float32"
+    device = "cpu"
+    mean = -0.260
+    std = 0.231
+    data = None
+
+
+class Program_weight_tensor_meta_L_self_modules_features_modules_11_modules_branch2_modules_1_modules_bn_buffers_running_var_:
+    name = "L_self_modules_features_modules_11_modules_branch2_modules_1_modules_bn_buffers_running_var_"
+    shape = [192]
+    dtype = "torch.float32"
+    device = "cpu"
+    mean = 0.266
+    std = 0.162
+    data = None
+
+
+class Program_weight_tensor_meta_L_self_modules_features_modules_11_modules_branch2_modules_1_modules_bn_parameters_weight_:
+    name = "L_self_modules_features_modules_11_modules_branch2_modules_1_modules_bn_parameters_weight_"
+    shape = [192]
+    dtype = "torch.float32"
+    device = "cpu"
+    mean = 1.000
+    std = 0.000
+    data = None
+
+
+class Program_weight_tensor_meta_L_self_modules_features_modules_11_modules_branch2_modules_1_modules_bn_parameters_bias_:
+    name = "L_self_modules_features_modules_11_modules_branch2_modules_1_modules_bn_parameters_bias_"
+    shape = [192]
+    dtype = "torch.float32"
+    device = "cpu"
+    mean = -0.725
+    std = 0.387
+    data = None
+
+
+class Program_weight_tensor_meta_L_self_modules_features_modules_11_modules_branch2_modules_2_modules_conv_parameters_weight_:
+    name = "L_self_modules_features_modules_11_modules_branch2_modules_2_modules_conv_parameters_weight_"
+    shape = [224, 192, 1, 7]
+    dtype = "torch.float32"
+    device = "cpu"
+    mean = -0.002
+    std = 0.018
+    data = None
+
+
+class Program_weight_tensor_meta_L_self_modules_features_modules_11_modules_branch2_modules_2_modules_bn_buffers_running_mean_:
+    name = "L_self_modules_features_modules_11_modules_branch2_modules_2_modules_bn_buffers_running_mean_"
+    shape = [224]
+    dtype = "torch.float32"
+    device = "cpu"
+    mean = -0.343
+    std = 0.309
+    data = None
+
+
+class Program_weight_tensor_meta_L_self_modules_features_modules_11_modules_branch2_modules_2_modules_bn_buffers_running_var_:
+    name = "L_self_modules_features_modules_11_modules_branch2_modules_2_modules_bn_buffers_running_var_"
+    shape = [224]
+    dtype = "torch.float32"
+    device = "cpu"
+    mean = 0.215
+    std = 0.107
+    data = None
+
+
+class Program_weight_tensor_meta_L_self_modules_features_modules_11_modules_branch2_modules_2_modules_bn_parameters_weight_:
+    name = "L_self_modules_features_modules_11_modules_branch2_modules_2_modules_bn_parameters_weight_"
+    shape = [224]
+    dtype = "torch.float32"
+    device = "cpu"
+    mean = 1.000
+    std = 0.000
+    data = None
+
+
+class Program_weight_tensor_meta_L_self_modules_features_modules_11_modules_branch2_modules_2_modules_bn_parameters_bias_:
+    name = "L_self_modules_features_modules_11_modules_branch2_modules_2_modules_bn_parameters_bias_"
+    shape = [224]
+    dtype = "torch.float32"
+    device = "cpu"
+    mean = -0.793
+    std = 0.309
+    data = None
+
+
+class Program_weight_tensor_meta_L_self_modules_features_modules_11_modules_branch2_modules_3_modules_conv_parameters_weight_:
+    name = "L_self_modules_features_modules_11_modules_branch2_modules_3_modules_conv_parameters_weight_"
+    shape = [224, 224, 7, 1]
+    dtype = "torch.float32"
+    device = "cpu"
+    mean = -0.001
+    std = 0.016
+    data = None
+
+
+class Program_weight_tensor_meta_L_self_modules_features_modules_11_modules_branch2_modules_3_modules_bn_buffers_running_mean_:
+    name = "L_self_modules_features_modules_11_modules_branch2_modules_3_modules_bn_buffers_running_mean_"
+    shape = [224]
+    dtype = "torch.float32"
+    device = "cpu"
+    mean = -0.270
+    std = 0.235
+    data = None
+
+
+class Program_weight_tensor_meta_L_self_modules_features_modules_11_modules_branch2_modules_3_modules_bn_buffers_running_var_:
+    name = "L_self_modules_features_modules_11_modules_branch2_modules_3_modules_bn_buffers_running_var_"
+    shape = [224]
+    dtype = "torch.float32"
+    device = "cpu"
+    mean = 0.217
+    std = 0.100
+    data = None
+
+
+class Program_weight_tensor_meta_L_self_modules_features_modules_11_modules_branch2_modules_3_modules_bn_parameters_weight_:
+    name = "L_self_modules_features_modules_11_modules_branch2_modules_3_modules_bn_parameters_weight_"
+    shape = [224]
+    dtype = "torch.float32"
+    device = "cpu"
+    mean = 1.000
+    std = 0.000
+    data = None
+
+
+class Program_weight_tensor_meta_L_self_modules_features_modules_11_modules_branch2_modules_3_modules_bn_parameters_bias_:
+    name = "L_self_modules_features_modules_11_modules_branch2_modules_3_modules_bn_parameters_bias_"
+    shape = [224]
+    dtype = "torch.float32"
+    device = "cpu"
+    mean = -0.609
+    std = 0.372
+    data = None
+
+
+class Program_weight_tensor_meta_L_self_modules_features_modules_11_modules_branch2_modules_4_modules_conv_parameters_weight_:
+    name = "L_self_modules_features_modules_11_modules_branch2_modules_4_modules_conv_parameters_weight_"
+    shape = [256, 224, 1, 7]
+    dtype = "torch.float32"
+    device = "cpu"
+    mean = -0.002
+    std = 0.015
+    data = None
+
+
+class Program_weight_tensor_meta_L_self_modules_features_modules_11_modules_branch2_modules_4_modules_bn_buffers_running_mean_:
+    name = "L_self_modules_features_modules_11_modules_branch2_modules_4_modules_bn_buffers_running_mean_"
+    shape = [256]
+    dtype = "torch.float32"
+    device = "cpu"
+    mean = -0.596
+    std = 0.303
+    data = None
+
+
+class Program_weight_tensor_meta_L_self_modules_features_modules_11_modules_branch2_modules_4_modules_bn_buffers_running_var_:
+    name = "L_self_modules_features_modules_11_modules_branch2_modules_4_modules_bn_buffers_running_var_"
+    shape = [256]
+    dtype = "torch.float32"
+    device = "cpu"
+    mean = 0.366
+    std = 0.293
+    data = None
+
+
+class Program_weight_tensor_meta_L_self_modules_features_modules_11_modules_branch2_modules_4_modules_bn_parameters_weight_:
+    name = "L_self_modules_features_modules_11_modules_branch2_modules_4_modules_bn_parameters_weight_"
+    shape = [256]
+    dtype = "torch.float32"
+    device = "cpu"
+    mean = 1.000
+    std = 0.000
+    data = None
+
+
+class Program_weight_tensor_meta_L_self_modules_features_modules_11_modules_branch2_modules_4_modules_bn_parameters_bias_:
+    name = "L_self_modules_features_modules_11_modules_branch2_modules_4_modules_bn_parameters_bias_"
+    shape = [256]
+    dtype = "torch.float32"
+    device = "cpu"
+    mean = -0.706
+    std = 0.342
+    data = None
+
+
+class Program_weight_tensor_meta_L_self_modules_features_modules_11_modules_branch3_modules_1_modules_conv_parameters_weight_:
+    name = "L_self_modules_features_modules_11_modules_branch3_modules_1_modules_conv_parameters_weight_"
+    shape = [128, 1024, 1, 1]
+    dtype = "torch.float32"
+    device = "cpu"
+    mean = -0.001
+    std = 0.027
+    data = None
+
+
+class Program_weight_tensor_meta_L_self_modules_features_modules_11_modules_branch3_modules_1_modules_bn_buffers_running_mean_:
+    name = "L_self_modules_features_modules_11_modules_branch3_modules_1_modules_bn_buffers_running_mean_"
+    shape = [128]
+    dtype = "torch.float32"
+    device = "cpu"
+    mean = -0.177
+    std = 0.446
+    data = None
+
+
+class Program_weight_tensor_meta_L_self_modules_features_modules_11_modules_branch3_modules_1_modules_bn_buffers_running_var_:
+    name = "L_self_modules_features_modules_11_modules_branch3_modules_1_modules_bn_buffers_running_var_"
+    shape = [128]
+    dtype = "torch.float32"
+    device = "cpu"
+    mean = 0.305
+    std = 0.134
+    data = None
+
+
+class Program_weight_tensor_meta_L_self_modules_features_modules_11_modules_branch3_modules_1_modules_bn_parameters_weight_:
+    name = "L_self_modules_features_modules_11_modules_branch3_modules_1_modules_bn_parameters_weight_"
+    shape = [128]
+    dtype = "torch.float32"
+    device = "cpu"
+    mean = 1.000
+    std = 0.000
+    data = None
+
+
+class Program_weight_tensor_meta_L_self_modules_features_modules_11_modules_branch3_modules_1_modules_bn_parameters_bias_:
+    name = "L_self_modules_features_modules_11_modules_branch3_modules_1_modules_bn_parameters_bias_"
+    shape = [128]
+    dtype = "torch.float32"
+    device = "cpu"
+    mean = -0.577
+    std = 0.168
+    data = None
+
+
+class Program_weight_tensor_meta_L_self_modules_features_modules_12_modules_branch0_modules_conv_parameters_weight_:
+    name = "L_self_modules_features_modules_12_modules_branch0_modules_conv_parameters_weight_"
+    shape = [384, 1024, 1, 1]
+    dtype = "torch.float32"
+    device = "cpu"
+    mean = -0.001
+    std = 0.031
+    data = None
+
+
+class Program_weight_tensor_meta_L_self_modules_features_modules_12_modules_branch0_modules_bn_buffers_running_mean_:
+    name = "L_self_modules_features_modules_12_modules_branch0_modules_bn_buffers_running_mean_"
+    shape = [384]
+    dtype = "torch.float32"
+    device = "cpu"
+    mean = -0.257
+    std = 0.425
+    data = None
+
+
+class Program_weight_tensor_meta_L_self_modules_features_modules_12_modules_branch0_modules_bn_buffers_running_var_:
+    name = "L_self_modules_features_modules_12_modules_branch0_modules_bn_buffers_running_var_"
+    shape = [384]
+    dtype = "torch.float32"
+    device = "cpu"
+    mean = 0.647
+    std = 0.639
+    data = None
+
+
+class Program_weight_tensor_meta_L_self_modules_features_modules_12_modules_branch0_modules_bn_parameters_weight_:
+    name = "L_self_modules_features_modules_12_modules_branch0_modules_bn_parameters_weight_"
+    shape = [384]
+    dtype = "torch.float32"
+    device = "cpu"
+    mean = 1.000
+    std = 0.000
+    data = None
+
+
+class Program_weight_tensor_meta_L_self_modules_features_modules_12_modules_branch0_modules_bn_parameters_bias_:
+    name = (
+        "L_self_modules_features_modules_12_modules_branch0_modules_bn_parameters_bias_"
+    )
+    shape = [384]
+    dtype = "torch.float32"
+    device = "cpu"
+    mean = -0.383
+    std = 0.426
+    data = None
+
+
+class Program_weight_tensor_meta_L_self_modules_features_modules_12_modules_branch1_modules_0_modules_conv_parameters_weight_:
+    name = "L_self_modules_features_modules_12_modules_branch1_modules_0_modules_conv_parameters_weight_"
+    shape = [192, 1024, 1, 1]
+    dtype = "torch.float32"
+    device = "cpu"
+    mean = -0.001
+    std = 0.022
+    data = None
+
+
+class Program_weight_tensor_meta_L_self_modules_features_modules_12_modules_branch1_modules_0_modules_bn_buffers_running_mean_:
+    name = "L_self_modules_features_modules_12_modules_branch1_modules_0_modules_bn_buffers_running_mean_"
+    shape = [192]
+    dtype = "torch.float32"
+    device = "cpu"
+    mean = -0.189
+    std = 0.353
+    data = None
+
+
+class Program_weight_tensor_meta_L_self_modules_features_modules_12_modules_branch1_modules_0_modules_bn_buffers_running_var_:
+    name = "L_self_modules_features_modules_12_modules_branch1_modules_0_modules_bn_buffers_running_var_"
+    shape = [192]
+    dtype = "torch.float32"
+    device = "cpu"
+    mean = 0.255
+    std = 0.115
+    data = None
+
+
+class Program_weight_tensor_meta_L_self_modules_features_modules_12_modules_branch1_modules_0_modules_bn_parameters_weight_:
+    name = "L_self_modules_features_modules_12_modules_branch1_modules_0_modules_bn_parameters_weight_"
+    shape = [192]
+    dtype = "torch.float32"
+    device = "cpu"
+    mean = 1.000
+    std = 0.000
+    data = None
+
+
+class Program_weight_tensor_meta_L_self_modules_features_modules_12_modules_branch1_modules_0_modules_bn_parameters_bias_:
+    name = "L_self_modules_features_modules_12_modules_branch1_modules_0_modules_bn_parameters_bias_"
+    shape = [192]
+    dtype = "torch.float32"
+    device = "cpu"
+    mean = -0.749
+    std = 0.239
+    data = None
+
+
+class Program_weight_tensor_meta_L_self_modules_features_modules_12_modules_branch1_modules_1_modules_conv_parameters_weight_:
+    name = "L_self_modules_features_modules_12_modules_branch1_modules_1_modules_conv_parameters_weight_"
+    shape = [224, 192, 1, 7]
+    dtype = "torch.float32"
+    device = "cpu"
+    mean = -0.001
+    std = 0.015
+    data = None
+
+
+class Program_weight_tensor_meta_L_self_modules_features_modules_12_modules_branch1_modules_1_modules_bn_buffers_running_mean_:
+    name = "L_self_modules_features_modules_12_modules_branch1_modules_1_modules_bn_buffers_running_mean_"
+    shape = [224]
+    dtype = "torch.float32"
+    device = "cpu"
+    mean = -0.170
+    std = 0.170
+    data = None
+
+
+class Program_weight_tensor_meta_L_self_modules_features_modules_12_modules_branch1_modules_1_modules_bn_buffers_running_var_:
+    name = "L_self_modules_features_modules_12_modules_branch1_modules_1_modules_bn_buffers_running_var_"
+    shape = [224]
+    dtype = "torch.float32"
+    device = "cpu"
+    mean = 0.190
+    std = 0.112
+    data = None
+
+
+class Program_weight_tensor_meta_L_self_modules_features_modules_12_modules_branch1_modules_1_modules_bn_parameters_weight_:
+    name = "L_self_modules_features_modules_12_modules_branch1_modules_1_modules_bn_parameters_weight_"
+    shape = [224]
+    dtype = "torch.float32"
+    device = "cpu"
+    mean = 1.000
+    std = 0.000
+    data = None
+
+
+class Program_weight_tensor_meta_L_self_modules_features_modules_12_modules_branch1_modules_1_modules_bn_parameters_bias_:
+    name = "L_self_modules_features_modules_12_modules_branch1_modules_1_modules_bn_parameters_bias_"
+    shape = [224]
+    dtype = "torch.float32"
+    device = "cpu"
+    mean = -0.344
+    std = 0.285
+    data = None
+
+
+class Program_weight_tensor_meta_L_self_modules_features_modules_12_modules_branch1_modules_2_modules_conv_parameters_weight_:
+    name = "L_self_modules_features_modules_12_modules_branch1_modules_2_modules_conv_parameters_weight_"
+    shape = [256, 224, 7, 1]
+    dtype = "torch.float32"
+    device = "cpu"
+    mean = -0.002
+    std = 0.014
+    data = None
+
+
+class Program_weight_tensor_meta_L_self_modules_features_modules_12_modules_branch1_modules_2_modules_bn_buffers_running_mean_:
+    name = "L_self_modules_features_modules_12_modules_branch1_modules_2_modules_bn_buffers_running_mean_"
+    shape = [256]
+    dtype = "torch.float32"
+    device = "cpu"
+    mean = -0.486
+    std = 0.320
+    data = None
+
+
+class Program_weight_tensor_meta_L_self_modules_features_modules_12_modules_branch1_modules_2_modules_bn_buffers_running_var_:
+    name = "L_self_modules_features_modules_12_modules_branch1_modules_2_modules_bn_buffers_running_var_"
+    shape = [256]
+    dtype = "torch.float32"
+    device = "cpu"
+    mean = 0.491
+    std = 0.383
+    data = None
+
+
+class Program_weight_tensor_meta_L_self_modules_features_modules_12_modules_branch1_modules_2_modules_bn_parameters_weight_:
+    name = "L_self_modules_features_modules_12_modules_branch1_modules_2_modules_bn_parameters_weight_"
+    shape = [256]
+    dtype = "torch.float32"
+    device = "cpu"
+    mean = 1.000
+    std = 0.000
+    data = None
+
+
+class Program_weight_tensor_meta_L_self_modules_features_modules_12_modules_branch1_modules_2_modules_bn_parameters_bias_:
+    name = "L_self_modules_features_modules_12_modules_branch1_modules_2_modules_bn_parameters_bias_"
+    shape = [256]
+    dtype = "torch.float32"
+    device = "cpu"
+    mean = -0.528
+    std = 0.256
+    data = None
+
+
+class Program_weight_tensor_meta_L_self_modules_features_modules_12_modules_branch2_modules_0_modules_conv_parameters_weight_:
+    name = "L_self_modules_features_modules_12_modules_branch2_modules_0_modules_conv_parameters_weight_"
+    shape = [192, 1024, 1, 1]
+    dtype = "torch.float32"
+    device = "cpu"
+    mean = -0.001
+    std = 0.020
+    data = None
+
+
+class Program_weight_tensor_meta_L_self_modules_features_modules_12_modules_branch2_modules_0_modules_bn_buffers_running_mean_:
+    name = "L_self_modules_features_modules_12_modules_branch2_modules_0_modules_bn_buffers_running_mean_"
+    shape = [192]
+    dtype = "torch.float32"
+    device = "cpu"
+    mean = -0.019
+    std = 0.540
+    data = None
+
+
+class Program_weight_tensor_meta_L_self_modules_features_modules_12_modules_branch2_modules_0_modules_bn_buffers_running_var_:
+    name = "L_self_modules_features_modules_12_modules_branch2_modules_0_modules_bn_buffers_running_var_"
+    shape = [192]
+    dtype = "torch.float32"
+    device = "cpu"
+    mean = 0.334
+    std = 0.268
+    data = None
+
+
+class Program_weight_tensor_meta_L_self_modules_features_modules_12_modules_branch2_modules_0_modules_bn_parameters_weight_:
+    name = "L_self_modules_features_modules_12_modules_branch2_modules_0_modules_bn_parameters_weight_"
+    shape = [192]
+    dtype = "torch.float32"
+    device = "cpu"
+    mean = 1.000
+    std = 0.000
+    data = None
+
+
+class Program_weight_tensor_meta_L_self_modules_features_modules_12_modules_branch2_modules_0_modules_bn_parameters_bias_:
+    name = "L_self_modules_features_modules_12_modules_branch2_modules_0_modules_bn_parameters_bias_"
+    shape = [192]
+    dtype = "torch.float32"
+    device = "cpu"
+    mean = -0.480
+    std = 0.395
+    data = None
+
+
+class Program_weight_tensor_meta_L_self_modules_features_modules_12_modules_branch2_modules_1_modules_conv_parameters_weight_:
+    name = "L_self_modules_features_modules_12_modules_branch2_modules_1_modules_conv_parameters_weight_"
+    shape = [192, 192, 7, 1]
+    dtype = "torch.float32"
+    device = "cpu"
+    mean = -0.001
+    std = 0.018
+    data = None
+
+
+class Program_weight_tensor_meta_L_self_modules_features_modules_12_modules_branch2_modules_1_modules_bn_buffers_running_mean_:
+    name = "L_self_modules_features_modules_12_modules_branch2_modules_1_modules_bn_buffers_running_mean_"
+    shape = [192]
+    dtype = "torch.float32"
+    device = "cpu"
+    mean = -0.253
+    std = 0.367
+    data = None
+
+
+class Program_weight_tensor_meta_L_self_modules_features_modules_12_modules_branch2_modules_1_modules_bn_buffers_running_var_:
+    name = "L_self_modules_features_modules_12_modules_branch2_modules_1_modules_bn_buffers_running_var_"
+    shape = [192]
+    dtype = "torch.float32"
+    device = "cpu"
+    mean = 0.460
+    std = 0.344
+    data = None
+
+
+class Program_weight_tensor_meta_L_self_modules_features_modules_12_modules_branch2_modules_1_modules_bn_parameters_weight_:
+    name = "L_self_modules_features_modules_12_modules_branch2_modules_1_modules_bn_parameters_weight_"
+    shape = [192]
+    dtype = "torch.float32"
+    device = "cpu"
+    mean = 1.000
+    std = 0.000
+    data = None
+
+
+class Program_weight_tensor_meta_L_self_modules_features_modules_12_modules_branch2_modules_1_modules_bn_parameters_bias_:
+    name = "L_self_modules_features_modules_12_modules_branch2_modules_1_modules_bn_parameters_bias_"
+    shape = [192]
+    dtype = "torch.float32"
+    device = "cpu"
+    mean = -0.600
+    std = 0.340
+    data = None
+
+
+class Program_weight_tensor_meta_L_self_modules_features_modules_12_modules_branch2_modules_2_modules_conv_parameters_weight_:
+    name = "L_self_modules_features_modules_12_modules_branch2_modules_2_modules_conv_parameters_weight_"
+    shape = [224, 192, 1, 7]
+    dtype = "torch.float32"
+    device = "cpu"
+    mean = -0.001
+    std = 0.016
+    data = None
+
+
+class Program_weight_tensor_meta_L_self_modules_features_modules_12_modules_branch2_modules_2_modules_bn_buffers_running_mean_:
+    name = "L_self_modules_features_modules_12_modules_branch2_modules_2_modules_bn_buffers_running_mean_"
+    shape = [224]
+    dtype = "torch.float32"
+    device = "cpu"
+    mean = -0.247
+    std = 0.296
+    data = None
+
+
+class Program_weight_tensor_meta_L_self_modules_features_modules_12_modules_branch2_modules_2_modules_bn_buffers_running_var_:
+    name = "L_self_modules_features_modules_12_modules_branch2_modules_2_modules_bn_buffers_running_var_"
+    shape = [224]
+    dtype = "torch.float32"
+    device = "cpu"
+    mean = 0.242
+    std = 0.115
+    data = None
+
+
+class Program_weight_tensor_meta_L_self_modules_features_modules_12_modules_branch2_modules_2_modules_bn_parameters_weight_:
+    name = "L_self_modules_features_modules_12_modules_branch2_modules_2_modules_bn_parameters_weight_"
+    shape = [224]
+    dtype = "torch.float32"
+    device = "cpu"
+    mean = 1.000
+    std = 0.000
+    data = None
+
+
+class Program_weight_tensor_meta_L_self_modules_features_modules_12_modules_branch2_modules_2_modules_bn_parameters_bias_:
+    name = "L_self_modules_features_modules_12_modules_branch2_modules_2_modules_bn_parameters_bias_"
+    shape = [224]
+    dtype = "torch.float32"
+    device = "cpu"
+    mean = -0.552
+    std = 0.301
+    data = None
+
+
+class Program_weight_tensor_meta_L_self_modules_features_modules_12_modules_branch2_modules_3_modules_conv_parameters_weight_:
+    name = "L_self_modules_features_modules_12_modules_branch2_modules_3_modules_conv_parameters_weight_"
+    shape = [224, 224, 7, 1]
+    dtype = "torch.float32"
+    device = "cpu"
+    mean = -0.001
+    std = 0.015
+    data = None
+
+
+class Program_weight_tensor_meta_L_self_modules_features_modules_12_modules_branch2_modules_3_modules_bn_buffers_running_mean_:
+    name = "L_self_modules_features_modules_12_modules_branch2_modules_3_modules_bn_buffers_running_mean_"
+    shape = [224]
+    dtype = "torch.float32"
+    device = "cpu"
+    mean = -0.235
+    std = 0.252
+    data = None
+
+
+class Program_weight_tensor_meta_L_self_modules_features_modules_12_modules_branch2_modules_3_modules_bn_buffers_running_var_:
+    name = "L_self_modules_features_modules_12_modules_branch2_modules_3_modules_bn_buffers_running_var_"
+    shape = [224]
+    dtype = "torch.float32"
+    device = "cpu"
+    mean = 0.295
+    std = 0.139
+    data = None
+
+
+class Program_weight_tensor_meta_L_self_modules_features_modules_12_modules_branch2_modules_3_modules_bn_parameters_weight_:
+    name = "L_self_modules_features_modules_12_modules_branch2_modules_3_modules_bn_parameters_weight_"
+    shape = [224]
+    dtype = "torch.float32"
+    device = "cpu"
+    mean = 1.000
+    std = 0.000
+    data = None
+
+
+class Program_weight_tensor_meta_L_self_modules_features_modules_12_modules_branch2_modules_3_modules_bn_parameters_bias_:
+    name = "L_self_modules_features_modules_12_modules_branch2_modules_3_modules_bn_parameters_bias_"
+    shape = [224]
+    dtype = "torch.float32"
+    device = "cpu"
+    mean = -0.510
+    std = 0.344
+    data = None
+
+
+class Program_weight_tensor_meta_L_self_modules_features_modules_12_modules_branch2_modules_4_modules_conv_parameters_weight_:
+    name = "L_self_modules_features_modules_12_modules_branch2_modules_4_modules_conv_parameters_weight_"
+    shape = [256, 224, 1, 7]
+    dtype = "torch.float32"
+    device = "cpu"
+    mean = -0.002
+    std = 0.014
+    data = None
+
+
+class Program_weight_tensor_meta_L_self_modules_features_modules_12_modules_branch2_modules_4_modules_bn_buffers_running_mean_:
+    name = "L_self_modules_features_modules_12_modules_branch2_modules_4_modules_bn_buffers_running_mean_"
+    shape = [256]
+    dtype = "torch.float32"
+    device = "cpu"
+    mean = -0.550
+    std = 0.310
+    data = None
+
+
+class Program_weight_tensor_meta_L_self_modules_features_modules_12_modules_branch2_modules_4_modules_bn_buffers_running_var_:
+    name = "L_self_modules_features_modules_12_modules_branch2_modules_4_modules_bn_buffers_running_var_"
+    shape = [256]
+    dtype = "torch.float32"
+    device = "cpu"
+    mean = 0.478
+    std = 0.441
+    data = None
+
+
+class Program_weight_tensor_meta_L_self_modules_features_modules_12_modules_branch2_modules_4_modules_bn_parameters_weight_:
+    name = "L_self_modules_features_modules_12_modules_branch2_modules_4_modules_bn_parameters_weight_"
+    shape = [256]
+    dtype = "torch.float32"
+    device = "cpu"
+    mean = 1.000
+    std = 0.000
+    data = None
+
+
+class Program_weight_tensor_meta_L_self_modules_features_modules_12_modules_branch2_modules_4_modules_bn_parameters_bias_:
+    name = "L_self_modules_features_modules_12_modules_branch2_modules_4_modules_bn_parameters_bias_"
+    shape = [256]
+    dtype = "torch.float32"
+    device = "cpu"
+    mean = -0.658
+    std = 0.330
+    data = None
+
+
+class Program_weight_tensor_meta_L_self_modules_features_modules_12_modules_branch3_modules_1_modules_conv_parameters_weight_:
+    name = "L_self_modules_features_modules_12_modules_branch3_modules_1_modules_conv_parameters_weight_"
+    shape = [128, 1024, 1, 1]
+    dtype = "torch.float32"
+    device = "cpu"
+    mean = -0.001
+    std = 0.028
+    data = None
+
+
+class Program_weight_tensor_meta_L_self_modules_features_modules_12_modules_branch3_modules_1_modules_bn_buffers_running_mean_:
+    name = "L_self_modules_features_modules_12_modules_branch3_modules_1_modules_bn_buffers_running_mean_"
+    shape = [128]
+    dtype = "torch.float32"
+    device = "cpu"
+    mean = -0.225
+    std = 0.263
+    data = None
+
+
+class Program_weight_tensor_meta_L_self_modules_features_modules_12_modules_branch3_modules_1_modules_bn_buffers_running_var_:
+    name = "L_self_modules_features_modules_12_modules_branch3_modules_1_modules_bn_buffers_running_var_"
+    shape = [128]
+    dtype = "torch.float32"
+    device = "cpu"
+    mean = 0.180
+    std = 0.061
+    data = None
+
+
+class Program_weight_tensor_meta_L_self_modules_features_modules_12_modules_branch3_modules_1_modules_bn_parameters_weight_:
+    name = "L_self_modules_features_modules_12_modules_branch3_modules_1_modules_bn_parameters_weight_"
+    shape = [128]
+    dtype = "torch.float32"
+    device = "cpu"
+    mean = 1.000
+    std = 0.000
+    data = None
+
+
+class Program_weight_tensor_meta_L_self_modules_features_modules_12_modules_branch3_modules_1_modules_bn_parameters_bias_:
+    name = "L_self_modules_features_modules_12_modules_branch3_modules_1_modules_bn_parameters_bias_"
+    shape = [128]
+    dtype = "torch.float32"
+    device = "cpu"
+    mean = -0.789
+    std = 0.177
+    data = None
+
+
+class Program_weight_tensor_meta_L_self_modules_features_modules_13_modules_branch0_modules_conv_parameters_weight_:
+    name = "L_self_modules_features_modules_13_modules_branch0_modules_conv_parameters_weight_"
+    shape = [384, 1024, 1, 1]
+    dtype = "torch.float32"
+    device = "cpu"
+    mean = -0.001
+    std = 0.028
+    data = None
+
+
+class Program_weight_tensor_meta_L_self_modules_features_modules_13_modules_branch0_modules_bn_buffers_running_mean_:
+    name = "L_self_modules_features_modules_13_modules_branch0_modules_bn_buffers_running_mean_"
+    shape = [384]
+    dtype = "torch.float32"
+    device = "cpu"
+    mean = -0.106
+    std = 0.373
+    data = None
+
+
+class Program_weight_tensor_meta_L_self_modules_features_modules_13_modules_branch0_modules_bn_buffers_running_var_:
+    name = "L_self_modules_features_modules_13_modules_branch0_modules_bn_buffers_running_var_"
+    shape = [384]
+    dtype = "torch.float32"
+    device = "cpu"
+    mean = 0.622
+    std = 0.548
+    data = None
+
+
+class Program_weight_tensor_meta_L_self_modules_features_modules_13_modules_branch0_modules_bn_parameters_weight_:
+    name = "L_self_modules_features_modules_13_modules_branch0_modules_bn_parameters_weight_"
+    shape = [384]
+    dtype = "torch.float32"
+    device = "cpu"
+    mean = 1.000
+    std = 0.000
+    data = None
+
+
+class Program_weight_tensor_meta_L_self_modules_features_modules_13_modules_branch0_modules_bn_parameters_bias_:
+    name = (
+        "L_self_modules_features_modules_13_modules_branch0_modules_bn_parameters_bias_"
+    )
+    shape = [384]
+    dtype = "torch.float32"
+    device = "cpu"
+    mean = -0.066
+    std = 0.541
+    data = None
+
+
+class Program_weight_tensor_meta_L_self_modules_features_modules_13_modules_branch1_modules_0_modules_conv_parameters_weight_:
+    name = "L_self_modules_features_modules_13_modules_branch1_modules_0_modules_conv_parameters_weight_"
+    shape = [192, 1024, 1, 1]
+    dtype = "torch.float32"
+    device = "cpu"
+    mean = -0.001
+    std = 0.022
+    data = None
+
+
+class Program_weight_tensor_meta_L_self_modules_features_modules_13_modules_branch1_modules_0_modules_bn_buffers_running_mean_:
+    name = "L_self_modules_features_modules_13_modules_branch1_modules_0_modules_bn_buffers_running_mean_"
+    shape = [192]
+    dtype = "torch.float32"
+    device = "cpu"
+    mean = -0.228
+    std = 0.298
+    data = None
+
+
+class Program_weight_tensor_meta_L_self_modules_features_modules_13_modules_branch1_modules_0_modules_bn_buffers_running_var_:
+    name = "L_self_modules_features_modules_13_modules_branch1_modules_0_modules_bn_buffers_running_var_"
+    shape = [192]
+    dtype = "torch.float32"
+    device = "cpu"
+    mean = 0.324
+    std = 0.179
+    data = None
+
+
+class Program_weight_tensor_meta_L_self_modules_features_modules_13_modules_branch1_modules_0_modules_bn_parameters_weight_:
+    name = "L_self_modules_features_modules_13_modules_branch1_modules_0_modules_bn_parameters_weight_"
+    shape = [192]
+    dtype = "torch.float32"
+    device = "cpu"
+    mean = 1.000
+    std = 0.000
+    data = None
+
+
+class Program_weight_tensor_meta_L_self_modules_features_modules_13_modules_branch1_modules_0_modules_bn_parameters_bias_:
+    name = "L_self_modules_features_modules_13_modules_branch1_modules_0_modules_bn_parameters_bias_"
+    shape = [192]
+    dtype = "torch.float32"
+    device = "cpu"
+    mean = -0.636
+    std = 0.287
+    data = None
+
+
+class Program_weight_tensor_meta_L_self_modules_features_modules_13_modules_branch1_modules_1_modules_conv_parameters_weight_:
+    name = "L_self_modules_features_modules_13_modules_branch1_modules_1_modules_conv_parameters_weight_"
+    shape = [224, 192, 1, 7]
+    dtype = "torch.float32"
+    device = "cpu"
+    mean = -0.001
+    std = 0.019
+    data = None
+
+
+class Program_weight_tensor_meta_L_self_modules_features_modules_13_modules_branch1_modules_1_modules_bn_buffers_running_mean_:
+    name = "L_self_modules_features_modules_13_modules_branch1_modules_1_modules_bn_buffers_running_mean_"
+    shape = [224]
+    dtype = "torch.float32"
+    device = "cpu"
+    mean = -0.302
+    std = 0.248
+    data = None
+
+
+class Program_weight_tensor_meta_L_self_modules_features_modules_13_modules_branch1_modules_1_modules_bn_buffers_running_var_:
+    name = "L_self_modules_features_modules_13_modules_branch1_modules_1_modules_bn_buffers_running_var_"
+    shape = [224]
+    dtype = "torch.float32"
+    device = "cpu"
+    mean = 0.282
+    std = 0.125
+    data = None
+
+
+class Program_weight_tensor_meta_L_self_modules_features_modules_13_modules_branch1_modules_1_modules_bn_parameters_weight_:
+    name = "L_self_modules_features_modules_13_modules_branch1_modules_1_modules_bn_parameters_weight_"
+    shape = [224]
+    dtype = "torch.float32"
+    device = "cpu"
+    mean = 1.000
+    std = 0.000
+    data = None
+
+
+class Program_weight_tensor_meta_L_self_modules_features_modules_13_modules_branch1_modules_1_modules_bn_parameters_bias_:
+    name = "L_self_modules_features_modules_13_modules_branch1_modules_1_modules_bn_parameters_bias_"
+    shape = [224]
+    dtype = "torch.float32"
+    device = "cpu"
+    mean = -0.620
+    std = 0.240
+    data = None
+
+
+class Program_weight_tensor_meta_L_self_modules_features_modules_13_modules_branch1_modules_2_modules_conv_parameters_weight_:
+    name = "L_self_modules_features_modules_13_modules_branch1_modules_2_modules_conv_parameters_weight_"
+    shape = [256, 224, 7, 1]
+    dtype = "torch.float32"
+    device = "cpu"
+    mean = -0.002
+    std = 0.018
+    data = None
+
+
+class Program_weight_tensor_meta_L_self_modules_features_modules_13_modules_branch1_modules_2_modules_bn_buffers_running_mean_:
+    name = "L_self_modules_features_modules_13_modules_branch1_modules_2_modules_bn_buffers_running_mean_"
+    shape = [256]
+    dtype = "torch.float32"
+    device = "cpu"
+    mean = -0.362
+    std = 0.222
+    data = None
+
+
+class Program_weight_tensor_meta_L_self_modules_features_modules_13_modules_branch1_modules_2_modules_bn_buffers_running_var_:
+    name = "L_self_modules_features_modules_13_modules_branch1_modules_2_modules_bn_buffers_running_var_"
+    shape = [256]
+    dtype = "torch.float32"
+    device = "cpu"
+    mean = 0.257
+    std = 0.067
+    data = None
+
+
+class Program_weight_tensor_meta_L_self_modules_features_modules_13_modules_branch1_modules_2_modules_bn_parameters_weight_:
+    name = "L_self_modules_features_modules_13_modules_branch1_modules_2_modules_bn_parameters_weight_"
+    shape = [256]
+    dtype = "torch.float32"
+    device = "cpu"
+    mean = 1.000
+    std = 0.000
+    data = None
+
+
+class Program_weight_tensor_meta_L_self_modules_features_modules_13_modules_branch1_modules_2_modules_bn_parameters_bias_:
+    name = "L_self_modules_features_modules_13_modules_branch1_modules_2_modules_bn_parameters_bias_"
+    shape = [256]
+    dtype = "torch.float32"
+    device = "cpu"
+    mean = -0.697
+    std = 0.203
+    data = None
+
+
+class Program_weight_tensor_meta_L_self_modules_features_modules_13_modules_branch2_modules_0_modules_conv_parameters_weight_:
+    name = "L_self_modules_features_modules_13_modules_branch2_modules_0_modules_conv_parameters_weight_"
+    shape = [192, 1024, 1, 1]
+    dtype = "torch.float32"
+    device = "cpu"
+    mean = -0.001
+    std = 0.020
+    data = None
+
+
+class Program_weight_tensor_meta_L_self_modules_features_modules_13_modules_branch2_modules_0_modules_bn_buffers_running_mean_:
+    name = "L_self_modules_features_modules_13_modules_branch2_modules_0_modules_bn_buffers_running_mean_"
+    shape = [192]
+    dtype = "torch.float32"
+    device = "cpu"
+    mean = -0.189
+    std = 0.283
+    data = None
+
+
+class Program_weight_tensor_meta_L_self_modules_features_modules_13_modules_branch2_modules_0_modules_bn_buffers_running_var_:
+    name = "L_self_modules_features_modules_13_modules_branch2_modules_0_modules_bn_buffers_running_var_"
+    shape = [192]
+    dtype = "torch.float32"
+    device = "cpu"
+    mean = 0.269
+    std = 0.173
+    data = None
+
+
+class Program_weight_tensor_meta_L_self_modules_features_modules_13_modules_branch2_modules_0_modules_bn_parameters_weight_:
+    name = "L_self_modules_features_modules_13_modules_branch2_modules_0_modules_bn_parameters_weight_"
+    shape = [192]
+    dtype = "torch.float32"
+    device = "cpu"
+    mean = 1.000
+    std = 0.000
+    data = None
+
+
+class Program_weight_tensor_meta_L_self_modules_features_modules_13_modules_branch2_modules_0_modules_bn_parameters_bias_:
+    name = "L_self_modules_features_modules_13_modules_branch2_modules_0_modules_bn_parameters_bias_"
+    shape = [192]
+    dtype = "torch.float32"
+    device = "cpu"
+    mean = -0.580
+    std = 0.315
+    data = None
+
+
+class Program_weight_tensor_meta_L_self_modules_features_modules_13_modules_branch2_modules_1_modules_conv_parameters_weight_:
+    name = "L_self_modules_features_modules_13_modules_branch2_modules_1_modules_conv_parameters_weight_"
+    shape = [192, 192, 7, 1]
+    dtype = "torch.float32"
+    device = "cpu"
+    mean = -0.001
+    std = 0.015
+    data = None
+
+
+class Program_weight_tensor_meta_L_self_modules_features_modules_13_modules_branch2_modules_1_modules_bn_buffers_running_mean_:
+    name = "L_self_modules_features_modules_13_modules_branch2_modules_1_modules_bn_buffers_running_mean_"
+    shape = [192]
+    dtype = "torch.float32"
+    device = "cpu"
+    mean = -0.210
+    std = 0.211
+    data = None
+
+
+class Program_weight_tensor_meta_L_self_modules_features_modules_13_modules_branch2_modules_1_modules_bn_buffers_running_var_:
+    name = "L_self_modules_features_modules_13_modules_branch2_modules_1_modules_bn_buffers_running_var_"
+    shape = [192]
+    dtype = "torch.float32"
+    device = "cpu"
+    mean = 0.228
+    std = 0.128
+    data = None
+
+
+class Program_weight_tensor_meta_L_self_modules_features_modules_13_modules_branch2_modules_1_modules_bn_parameters_weight_:
+    name = "L_self_modules_features_modules_13_modules_branch2_modules_1_modules_bn_parameters_weight_"
+    shape = [192]
+    dtype = "torch.float32"
+    device = "cpu"
+    mean = 1.000
+    std = 0.000
+    data = None
+
+
+class Program_weight_tensor_meta_L_self_modules_features_modules_13_modules_branch2_modules_1_modules_bn_parameters_bias_:
+    name = "L_self_modules_features_modules_13_modules_branch2_modules_1_modules_bn_parameters_bias_"
+    shape = [192]
+    dtype = "torch.float32"
+    device = "cpu"
+    mean = -0.348
+    std = 0.347
+    data = None
+
+
+class Program_weight_tensor_meta_L_self_modules_features_modules_13_modules_branch2_modules_2_modules_conv_parameters_weight_:
+    name = "L_self_modules_features_modules_13_modules_branch2_modules_2_modules_conv_parameters_weight_"
+    shape = [224, 192, 1, 7]
+    dtype = "torch.float32"
+    device = "cpu"
+    mean = -0.001
+    std = 0.013
+    data = None
+
+
+class Program_weight_tensor_meta_L_self_modules_features_modules_13_modules_branch2_modules_2_modules_bn_buffers_running_mean_:
+    name = "L_self_modules_features_modules_13_modules_branch2_modules_2_modules_bn_buffers_running_mean_"
+    shape = [224]
+    dtype = "torch.float32"
+    device = "cpu"
+    mean = -0.252
+    std = 0.298
+    data = None
+
+
+class Program_weight_tensor_meta_L_self_modules_features_modules_13_modules_branch2_modules_2_modules_bn_buffers_running_var_:
+    name = "L_self_modules_features_modules_13_modules_branch2_modules_2_modules_bn_buffers_running_var_"
+    shape = [224]
+    dtype = "torch.float32"
+    device = "cpu"
+    mean = 0.250
+    std = 0.114
+    data = None
+
+
+class Program_weight_tensor_meta_L_self_modules_features_modules_13_modules_branch2_modules_2_modules_bn_parameters_weight_:
+    name = "L_self_modules_features_modules_13_modules_branch2_modules_2_modules_bn_parameters_weight_"
+    shape = [224]
+    dtype = "torch.float32"
+    device = "cpu"
+    mean = 1.000
+    std = 0.000
+    data = None
+
+
+class Program_weight_tensor_meta_L_self_modules_features_modules_13_modules_branch2_modules_2_modules_bn_parameters_bias_:
+    name = "L_self_modules_features_modules_13_modules_branch2_modules_2_modules_bn_parameters_bias_"
+    shape = [224]
+    dtype = "torch.float32"
+    device = "cpu"
+    mean = -0.351
+    std = 0.211
+    data = None
+
+
+class Program_weight_tensor_meta_L_self_modules_features_modules_13_modules_branch2_modules_3_modules_conv_parameters_weight_:
+    name = "L_self_modules_features_modules_13_modules_branch2_modules_3_modules_conv_parameters_weight_"
+    shape = [224, 224, 7, 1]
+    dtype = "torch.float32"
+    device = "cpu"
+    mean = -0.000
+    std = 0.011
+    data = None
+
+
+class Program_weight_tensor_meta_L_self_modules_features_modules_13_modules_branch2_modules_3_modules_bn_buffers_running_mean_:
+    name = "L_self_modules_features_modules_13_modules_branch2_modules_3_modules_bn_buffers_running_mean_"
+    shape = [224]
+    dtype = "torch.float32"
+    device = "cpu"
+    mean = -0.157
+    std = 0.269
+    data = None
+
+
+class Program_weight_tensor_meta_L_self_modules_features_modules_13_modules_branch2_modules_3_modules_bn_buffers_running_var_:
+    name = "L_self_modules_features_modules_13_modules_branch2_modules_3_modules_bn_buffers_running_var_"
+    shape = [224]
+    dtype = "torch.float32"
+    device = "cpu"
+    mean = 0.267
+    std = 0.178
+    data = None
+
+
+class Program_weight_tensor_meta_L_self_modules_features_modules_13_modules_branch2_modules_3_modules_bn_parameters_weight_:
+    name = "L_self_modules_features_modules_13_modules_branch2_modules_3_modules_bn_parameters_weight_"
+    shape = [224]
+    dtype = "torch.float32"
+    device = "cpu"
+    mean = 1.000
+    std = 0.000
+    data = None
+
+
+class Program_weight_tensor_meta_L_self_modules_features_modules_13_modules_branch2_modules_3_modules_bn_parameters_bias_:
+    name = "L_self_modules_features_modules_13_modules_branch2_modules_3_modules_bn_parameters_bias_"
+    shape = [224]
+    dtype = "torch.float32"
+    device = "cpu"
+    mean = -0.286
+    std = 0.317
+    data = None
+
+
+class Program_weight_tensor_meta_L_self_modules_features_modules_13_modules_branch2_modules_4_modules_conv_parameters_weight_:
+    name = "L_self_modules_features_modules_13_modules_branch2_modules_4_modules_conv_parameters_weight_"
+    shape = [256, 224, 1, 7]
+    dtype = "torch.float32"
+    device = "cpu"
+    mean = -0.001
+    std = 0.010
+    data = None
+
+
+class Program_weight_tensor_meta_L_self_modules_features_modules_13_modules_branch2_modules_4_modules_bn_buffers_running_mean_:
+    name = "L_self_modules_features_modules_13_modules_branch2_modules_4_modules_bn_buffers_running_mean_"
+    shape = [256]
+    dtype = "torch.float32"
+    device = "cpu"
+    mean = -0.441
+    std = 0.441
+    data = None
+
+
+class Program_weight_tensor_meta_L_self_modules_features_modules_13_modules_branch2_modules_4_modules_bn_buffers_running_var_:
+    name = "L_self_modules_features_modules_13_modules_branch2_modules_4_modules_bn_buffers_running_var_"
+    shape = [256]
+    dtype = "torch.float32"
+    device = "cpu"
+    mean = 0.537
+    std = 0.496
+    data = None
+
+
+class Program_weight_tensor_meta_L_self_modules_features_modules_13_modules_branch2_modules_4_modules_bn_parameters_weight_:
+    name = "L_self_modules_features_modules_13_modules_branch2_modules_4_modules_bn_parameters_weight_"
+    shape = [256]
+    dtype = "torch.float32"
+    device = "cpu"
+    mean = 1.000
+    std = 0.000
+    data = None
+
+
+class Program_weight_tensor_meta_L_self_modules_features_modules_13_modules_branch2_modules_4_modules_bn_parameters_bias_:
+    name = "L_self_modules_features_modules_13_modules_branch2_modules_4_modules_bn_parameters_bias_"
+    shape = [256]
+    dtype = "torch.float32"
+    device = "cpu"
+    mean = -0.462
+    std = 0.262
+    data = None
+
+
+class Program_weight_tensor_meta_L_self_modules_features_modules_13_modules_branch3_modules_1_modules_conv_parameters_weight_:
+    name = "L_self_modules_features_modules_13_modules_branch3_modules_1_modules_conv_parameters_weight_"
+    shape = [128, 1024, 1, 1]
+    dtype = "torch.float32"
+    device = "cpu"
+    mean = -0.001
+    std = 0.027
+    data = None
+
+
+class Program_weight_tensor_meta_L_self_modules_features_modules_13_modules_branch3_modules_1_modules_bn_buffers_running_mean_:
+    name = "L_self_modules_features_modules_13_modules_branch3_modules_1_modules_bn_buffers_running_mean_"
+    shape = [128]
+    dtype = "torch.float32"
+    device = "cpu"
+    mean = -0.185
+    std = 0.230
+    data = None
+
+
+class Program_weight_tensor_meta_L_self_modules_features_modules_13_modules_branch3_modules_1_modules_bn_buffers_running_var_:
+    name = "L_self_modules_features_modules_13_modules_branch3_modules_1_modules_bn_buffers_running_var_"
+    shape = [128]
+    dtype = "torch.float32"
+    device = "cpu"
+    mean = 0.156
+    std = 0.053
+    data = None
+
+
+class Program_weight_tensor_meta_L_self_modules_features_modules_13_modules_branch3_modules_1_modules_bn_parameters_weight_:
+    name = "L_self_modules_features_modules_13_modules_branch3_modules_1_modules_bn_parameters_weight_"
+    shape = [128]
+    dtype = "torch.float32"
+    device = "cpu"
+    mean = 1.000
+    std = 0.000
+    data = None
+
+
+class Program_weight_tensor_meta_L_self_modules_features_modules_13_modules_branch3_modules_1_modules_bn_parameters_bias_:
+    name = "L_self_modules_features_modules_13_modules_branch3_modules_1_modules_bn_parameters_bias_"
+    shape = [128]
+    dtype = "torch.float32"
+    device = "cpu"
+    mean = -0.592
+    std = 0.168
+    data = None
+
+
+class Program_weight_tensor_meta_L_self_modules_features_modules_14_modules_branch0_modules_conv_parameters_weight_:
+    name = "L_self_modules_features_modules_14_modules_branch0_modules_conv_parameters_weight_"
+    shape = [384, 1024, 1, 1]
+    dtype = "torch.float32"
+    device = "cpu"
+    mean = -0.001
+    std = 0.027
+    data = None
+
+
+class Program_weight_tensor_meta_L_self_modules_features_modules_14_modules_branch0_modules_bn_buffers_running_mean_:
+    name = "L_self_modules_features_modules_14_modules_branch0_modules_bn_buffers_running_mean_"
+    shape = [384]
+    dtype = "torch.float32"
+    device = "cpu"
+    mean = -0.155
+    std = 0.500
+    data = None
+
+
+class Program_weight_tensor_meta_L_self_modules_features_modules_14_modules_branch0_modules_bn_buffers_running_var_:
+    name = "L_self_modules_features_modules_14_modules_branch0_modules_bn_buffers_running_var_"
+    shape = [384]
+    dtype = "torch.float32"
+    device = "cpu"
+    mean = 0.722
+    std = 0.551
+    data = None
+
+
+class Program_weight_tensor_meta_L_self_modules_features_modules_14_modules_branch0_modules_bn_parameters_weight_:
+    name = "L_self_modules_features_modules_14_modules_branch0_modules_bn_parameters_weight_"
+    shape = [384]
+    dtype = "torch.float32"
+    device = "cpu"
+    mean = 1.000
+    std = 0.000
+    data = None
+
+
+class Program_weight_tensor_meta_L_self_modules_features_modules_14_modules_branch0_modules_bn_parameters_bias_:
+    name = (
+        "L_self_modules_features_modules_14_modules_branch0_modules_bn_parameters_bias_"
+    )
+    shape = [384]
+    dtype = "torch.float32"
+    device = "cpu"
+    mean = 0.104
+    std = 0.586
+    data = None
+
+
+class Program_weight_tensor_meta_L_self_modules_features_modules_14_modules_branch1_modules_0_modules_conv_parameters_weight_:
+    name = "L_self_modules_features_modules_14_modules_branch1_modules_0_modules_conv_parameters_weight_"
+    shape = [192, 1024, 1, 1]
+    dtype = "torch.float32"
+    device = "cpu"
+    mean = -0.001
+    std = 0.023
+    data = None
+
+
+class Program_weight_tensor_meta_L_self_modules_features_modules_14_modules_branch1_modules_0_modules_bn_buffers_running_mean_:
+    name = "L_self_modules_features_modules_14_modules_branch1_modules_0_modules_bn_buffers_running_mean_"
+    shape = [192]
+    dtype = "torch.float32"
+    device = "cpu"
+    mean = -0.188
+    std = 0.367
+    data = None
+
+
+class Program_weight_tensor_meta_L_self_modules_features_modules_14_modules_branch1_modules_0_modules_bn_buffers_running_var_:
+    name = "L_self_modules_features_modules_14_modules_branch1_modules_0_modules_bn_buffers_running_var_"
+    shape = [192]
+    dtype = "torch.float32"
+    device = "cpu"
+    mean = 0.413
+    std = 0.221
+    data = None
+
+
+class Program_weight_tensor_meta_L_self_modules_features_modules_14_modules_branch1_modules_0_modules_bn_parameters_weight_:
+    name = "L_self_modules_features_modules_14_modules_branch1_modules_0_modules_bn_parameters_weight_"
+    shape = [192]
+    dtype = "torch.float32"
+    device = "cpu"
+    mean = 1.000
+    std = 0.000
+    data = None
+
+
+class Program_weight_tensor_meta_L_self_modules_features_modules_14_modules_branch1_modules_0_modules_bn_parameters_bias_:
+    name = "L_self_modules_features_modules_14_modules_branch1_modules_0_modules_bn_parameters_bias_"
+    shape = [192]
+    dtype = "torch.float32"
+    device = "cpu"
+    mean = -0.620
+    std = 0.263
+    data = None
+
+
+class Program_weight_tensor_meta_L_self_modules_features_modules_14_modules_branch1_modules_1_modules_conv_parameters_weight_:
+    name = "L_self_modules_features_modules_14_modules_branch1_modules_1_modules_conv_parameters_weight_"
+    shape = [224, 192, 1, 7]
+    dtype = "torch.float32"
+    device = "cpu"
+    mean = -0.001
+    std = 0.018
+    data = None
+
+
+class Program_weight_tensor_meta_L_self_modules_features_modules_14_modules_branch1_modules_1_modules_bn_buffers_running_mean_:
+    name = "L_self_modules_features_modules_14_modules_branch1_modules_1_modules_bn_buffers_running_mean_"
+    shape = [224]
+    dtype = "torch.float32"
+    device = "cpu"
+    mean = -0.256
+    std = 0.227
+    data = None
+
+
+class Program_weight_tensor_meta_L_self_modules_features_modules_14_modules_branch1_modules_1_modules_bn_buffers_running_var_:
+    name = "L_self_modules_features_modules_14_modules_branch1_modules_1_modules_bn_buffers_running_var_"
+    shape = [224]
+    dtype = "torch.float32"
+    device = "cpu"
+    mean = 0.282
+    std = 0.194
+    data = None
+
+
+class Program_weight_tensor_meta_L_self_modules_features_modules_14_modules_branch1_modules_1_modules_bn_parameters_weight_:
+    name = "L_self_modules_features_modules_14_modules_branch1_modules_1_modules_bn_parameters_weight_"
+    shape = [224]
+    dtype = "torch.float32"
+    device = "cpu"
+    mean = 1.000
+    std = 0.000
+    data = None
+
+
+class Program_weight_tensor_meta_L_self_modules_features_modules_14_modules_branch1_modules_1_modules_bn_parameters_bias_:
+    name = "L_self_modules_features_modules_14_modules_branch1_modules_1_modules_bn_parameters_bias_"
+    shape = [224]
+    dtype = "torch.float32"
+    device = "cpu"
+    mean = -0.566
+    std = 0.286
+    data = None
+
+
+class Program_weight_tensor_meta_L_self_modules_features_modules_14_modules_branch1_modules_2_modules_conv_parameters_weight_:
+    name = "L_self_modules_features_modules_14_modules_branch1_modules_2_modules_conv_parameters_weight_"
+    shape = [256, 224, 7, 1]
+    dtype = "torch.float32"
+    device = "cpu"
+    mean = -0.001
+    std = 0.017
+    data = None
+
+
+class Program_weight_tensor_meta_L_self_modules_features_modules_14_modules_branch1_modules_2_modules_bn_buffers_running_mean_:
+    name = "L_self_modules_features_modules_14_modules_branch1_modules_2_modules_bn_buffers_running_mean_"
+    shape = [256]
+    dtype = "torch.float32"
+    device = "cpu"
+    mean = -0.350
+    std = 0.273
+    data = None
+
+
+class Program_weight_tensor_meta_L_self_modules_features_modules_14_modules_branch1_modules_2_modules_bn_buffers_running_var_:
+    name = "L_self_modules_features_modules_14_modules_branch1_modules_2_modules_bn_buffers_running_var_"
+    shape = [256]
+    dtype = "torch.float32"
+    device = "cpu"
+    mean = 0.242
+    std = 0.124
+    data = None
+
+
+class Program_weight_tensor_meta_L_self_modules_features_modules_14_modules_branch1_modules_2_modules_bn_parameters_weight_:
+    name = "L_self_modules_features_modules_14_modules_branch1_modules_2_modules_bn_parameters_weight_"
+    shape = [256]
+    dtype = "torch.float32"
+    device = "cpu"
+    mean = 1.000
+    std = 0.000
+    data = None
+
+
+class Program_weight_tensor_meta_L_self_modules_features_modules_14_modules_branch1_modules_2_modules_bn_parameters_bias_:
+    name = "L_self_modules_features_modules_14_modules_branch1_modules_2_modules_bn_parameters_bias_"
+    shape = [256]
+    dtype = "torch.float32"
+    device = "cpu"
+    mean = -0.608
+    std = 0.255
+    data = None
+
+
+class Program_weight_tensor_meta_L_self_modules_features_modules_14_modules_branch2_modules_0_modules_conv_parameters_weight_:
+    name = "L_self_modules_features_modules_14_modules_branch2_modules_0_modules_conv_parameters_weight_"
+    shape = [192, 1024, 1, 1]
+    dtype = "torch.float32"
+    device = "cpu"
+    mean = -0.000
+    std = 0.016
+    data = None
+
+
+class Program_weight_tensor_meta_L_self_modules_features_modules_14_modules_branch2_modules_0_modules_bn_buffers_running_mean_:
+    name = "L_self_modules_features_modules_14_modules_branch2_modules_0_modules_bn_buffers_running_mean_"
+    shape = [192]
+    dtype = "torch.float32"
+    device = "cpu"
+    mean = -0.105
+    std = 0.275
+    data = None
+
+
+class Program_weight_tensor_meta_L_self_modules_features_modules_14_modules_branch2_modules_0_modules_bn_buffers_running_var_:
+    name = "L_self_modules_features_modules_14_modules_branch2_modules_0_modules_bn_buffers_running_var_"
+    shape = [192]
+    dtype = "torch.float32"
+    device = "cpu"
+    mean = 0.228
+    std = 0.151
+    data = None
+
+
+class Program_weight_tensor_meta_L_self_modules_features_modules_14_modules_branch2_modules_0_modules_bn_parameters_weight_:
+    name = "L_self_modules_features_modules_14_modules_branch2_modules_0_modules_bn_parameters_weight_"
+    shape = [192]
+    dtype = "torch.float32"
+    device = "cpu"
+    mean = 1.000
+    std = 0.000
+    data = None
+
+
+class Program_weight_tensor_meta_L_self_modules_features_modules_14_modules_branch2_modules_0_modules_bn_parameters_bias_:
+    name = "L_self_modules_features_modules_14_modules_branch2_modules_0_modules_bn_parameters_bias_"
+    shape = [192]
+    dtype = "torch.float32"
+    device = "cpu"
+    mean = -0.440
+    std = 0.349
+    data = None
+
+
+class Program_weight_tensor_meta_L_self_modules_features_modules_14_modules_branch2_modules_1_modules_conv_parameters_weight_:
+    name = "L_self_modules_features_modules_14_modules_branch2_modules_1_modules_conv_parameters_weight_"
+    shape = [192, 192, 7, 1]
+    dtype = "torch.float32"
+    device = "cpu"
+    mean = -0.000
+    std = 0.011
+    data = None
+
+
+class Program_weight_tensor_meta_L_self_modules_features_modules_14_modules_branch2_modules_1_modules_bn_buffers_running_mean_:
+    name = "L_self_modules_features_modules_14_modules_branch2_modules_1_modules_bn_buffers_running_mean_"
+    shape = [192]
+    dtype = "torch.float32"
+    device = "cpu"
+    mean = -0.080
+    std = 0.257
+    data = None
+
+
+class Program_weight_tensor_meta_L_self_modules_features_modules_14_modules_branch2_modules_1_modules_bn_buffers_running_var_:
+    name = "L_self_modules_features_modules_14_modules_branch2_modules_1_modules_bn_buffers_running_var_"
+    shape = [192]
+    dtype = "torch.float32"
+    device = "cpu"
+    mean = 0.179
+    std = 0.134
+    data = None
+
+
+class Program_weight_tensor_meta_L_self_modules_features_modules_14_modules_branch2_modules_1_modules_bn_parameters_weight_:
+    name = "L_self_modules_features_modules_14_modules_branch2_modules_1_modules_bn_parameters_weight_"
+    shape = [192]
+    dtype = "torch.float32"
+    device = "cpu"
+    mean = 1.000
+    std = 0.000
+    data = None
+
+
+class Program_weight_tensor_meta_L_self_modules_features_modules_14_modules_branch2_modules_1_modules_bn_parameters_bias_:
+    name = "L_self_modules_features_modules_14_modules_branch2_modules_1_modules_bn_parameters_bias_"
+    shape = [192]
+    dtype = "torch.float32"
+    device = "cpu"
+    mean = -0.213
+    std = 0.324
+    data = None
+
+
+class Program_weight_tensor_meta_L_self_modules_features_modules_14_modules_branch2_modules_2_modules_conv_parameters_weight_:
+    name = "L_self_modules_features_modules_14_modules_branch2_modules_2_modules_conv_parameters_weight_"
+    shape = [224, 192, 1, 7]
+    dtype = "torch.float32"
+    device = "cpu"
+    mean = -0.000
+    std = 0.009
+    data = None
+
+
+class Program_weight_tensor_meta_L_self_modules_features_modules_14_modules_branch2_modules_2_modules_bn_buffers_running_mean_:
+    name = "L_self_modules_features_modules_14_modules_branch2_modules_2_modules_bn_buffers_running_mean_"
+    shape = [224]
+    dtype = "torch.float32"
+    device = "cpu"
+    mean = -0.126
+    std = 0.358
+    data = None
+
+
+class Program_weight_tensor_meta_L_self_modules_features_modules_14_modules_branch2_modules_2_modules_bn_buffers_running_var_:
+    name = "L_self_modules_features_modules_14_modules_branch2_modules_2_modules_bn_buffers_running_var_"
+    shape = [224]
+    dtype = "torch.float32"
+    device = "cpu"
+    mean = 0.230
+    std = 0.143
+    data = None
+
+
+class Program_weight_tensor_meta_L_self_modules_features_modules_14_modules_branch2_modules_2_modules_bn_parameters_weight_:
+    name = "L_self_modules_features_modules_14_modules_branch2_modules_2_modules_bn_parameters_weight_"
+    shape = [224]
+    dtype = "torch.float32"
+    device = "cpu"
+    mean = 1.000
+    std = 0.000
+    data = None
+
+
+class Program_weight_tensor_meta_L_self_modules_features_modules_14_modules_branch2_modules_2_modules_bn_parameters_bias_:
+    name = "L_self_modules_features_modules_14_modules_branch2_modules_2_modules_bn_parameters_bias_"
+    shape = [224]
+    dtype = "torch.float32"
+    device = "cpu"
+    mean = -0.203
+    std = 0.194
+    data = None
+
+
+class Program_weight_tensor_meta_L_self_modules_features_modules_14_modules_branch2_modules_3_modules_conv_parameters_weight_:
+    name = "L_self_modules_features_modules_14_modules_branch2_modules_3_modules_conv_parameters_weight_"
+    shape = [224, 224, 7, 1]
+    dtype = "torch.float32"
+    device = "cpu"
+    mean = -0.000
+    std = 0.008
+    data = None
+
+
+class Program_weight_tensor_meta_L_self_modules_features_modules_14_modules_branch2_modules_3_modules_bn_buffers_running_mean_:
+    name = "L_self_modules_features_modules_14_modules_branch2_modules_3_modules_bn_buffers_running_mean_"
+    shape = [224]
+    dtype = "torch.float32"
+    device = "cpu"
+    mean = -0.112
+    std = 0.311
+    data = None
+
+
+class Program_weight_tensor_meta_L_self_modules_features_modules_14_modules_branch2_modules_3_modules_bn_buffers_running_var_:
+    name = "L_self_modules_features_modules_14_modules_branch2_modules_3_modules_bn_buffers_running_var_"
+    shape = [224]
+    dtype = "torch.float32"
+    device = "cpu"
+    mean = 0.311
+    std = 0.204
+    data = None
+
+
+class Program_weight_tensor_meta_L_self_modules_features_modules_14_modules_branch2_modules_3_modules_bn_parameters_weight_:
+    name = "L_self_modules_features_modules_14_modules_branch2_modules_3_modules_bn_parameters_weight_"
+    shape = [224]
+    dtype = "torch.float32"
+    device = "cpu"
+    mean = 1.000
+    std = 0.000
+    data = None
+
+
+class Program_weight_tensor_meta_L_self_modules_features_modules_14_modules_branch2_modules_3_modules_bn_parameters_bias_:
+    name = "L_self_modules_features_modules_14_modules_branch2_modules_3_modules_bn_parameters_bias_"
+    shape = [224]
+    dtype = "torch.float32"
+    device = "cpu"
+    mean = -0.229
+    std = 0.269
+    data = None
+
+
+class Program_weight_tensor_meta_L_self_modules_features_modules_14_modules_branch2_modules_4_modules_conv_parameters_weight_:
+    name = "L_self_modules_features_modules_14_modules_branch2_modules_4_modules_conv_parameters_weight_"
+    shape = [256, 224, 1, 7]
+    dtype = "torch.float32"
+    device = "cpu"
+    mean = -0.001
+    std = 0.009
+    data = None
+
+
+class Program_weight_tensor_meta_L_self_modules_features_modules_14_modules_branch2_modules_4_modules_bn_buffers_running_mean_:
+    name = "L_self_modules_features_modules_14_modules_branch2_modules_4_modules_bn_buffers_running_mean_"
+    shape = [256]
+    dtype = "torch.float32"
+    device = "cpu"
+    mean = -0.394
+    std = 0.369
+    data = None
+
+
+class Program_weight_tensor_meta_L_self_modules_features_modules_14_modules_branch2_modules_4_modules_bn_buffers_running_var_:
+    name = "L_self_modules_features_modules_14_modules_branch2_modules_4_modules_bn_buffers_running_var_"
+    shape = [256]
+    dtype = "torch.float32"
+    device = "cpu"
+    mean = 0.571
+    std = 0.499
+    data = None
+
+
+class Program_weight_tensor_meta_L_self_modules_features_modules_14_modules_branch2_modules_4_modules_bn_parameters_weight_:
+    name = "L_self_modules_features_modules_14_modules_branch2_modules_4_modules_bn_parameters_weight_"
+    shape = [256]
+    dtype = "torch.float32"
+    device = "cpu"
+    mean = 1.000
+    std = 0.000
+    data = None
+
+
+class Program_weight_tensor_meta_L_self_modules_features_modules_14_modules_branch2_modules_4_modules_bn_parameters_bias_:
+    name = "L_self_modules_features_modules_14_modules_branch2_modules_4_modules_bn_parameters_bias_"
+    shape = [256]
+    dtype = "torch.float32"
+    device = "cpu"
+    mean = -0.495
+    std = 0.230
+    data = None
+
+
+class Program_weight_tensor_meta_L_self_modules_features_modules_14_modules_branch3_modules_1_modules_conv_parameters_weight_:
+    name = "L_self_modules_features_modules_14_modules_branch3_modules_1_modules_conv_parameters_weight_"
+    shape = [128, 1024, 1, 1]
+    dtype = "torch.float32"
+    device = "cpu"
+    mean = -0.001
+    std = 0.027
+    data = None
+
+
+class Program_weight_tensor_meta_L_self_modules_features_modules_14_modules_branch3_modules_1_modules_bn_buffers_running_mean_:
+    name = "L_self_modules_features_modules_14_modules_branch3_modules_1_modules_bn_buffers_running_mean_"
+    shape = [128]
+    dtype = "torch.float32"
+    device = "cpu"
+    mean = -0.046
+    std = 0.300
+    data = None
+
+
+class Program_weight_tensor_meta_L_self_modules_features_modules_14_modules_branch3_modules_1_modules_bn_buffers_running_var_:
+    name = "L_self_modules_features_modules_14_modules_branch3_modules_1_modules_bn_buffers_running_var_"
+    shape = [128]
+    dtype = "torch.float32"
+    device = "cpu"
+    mean = 0.192
+    std = 0.065
+    data = None
+
+
+class Program_weight_tensor_meta_L_self_modules_features_modules_14_modules_branch3_modules_1_modules_bn_parameters_weight_:
+    name = "L_self_modules_features_modules_14_modules_branch3_modules_1_modules_bn_parameters_weight_"
+    shape = [128]
+    dtype = "torch.float32"
+    device = "cpu"
+    mean = 1.000
+    std = 0.000
+    data = None
+
+
+class Program_weight_tensor_meta_L_self_modules_features_modules_14_modules_branch3_modules_1_modules_bn_parameters_bias_:
+    name = "L_self_modules_features_modules_14_modules_branch3_modules_1_modules_bn_parameters_bias_"
+    shape = [128]
+    dtype = "torch.float32"
+    device = "cpu"
+    mean = -0.442
+    std = 0.154
+    data = None
+
+
+class Program_weight_tensor_meta_L_self_modules_features_modules_15_modules_branch0_modules_conv_parameters_weight_:
+    name = "L_self_modules_features_modules_15_modules_branch0_modules_conv_parameters_weight_"
+    shape = [384, 1024, 1, 1]
+    dtype = "torch.float32"
+    device = "cpu"
+    mean = -0.001
+    std = 0.028
+    data = None
+
+
+class Program_weight_tensor_meta_L_self_modules_features_modules_15_modules_branch0_modules_bn_buffers_running_mean_:
+    name = "L_self_modules_features_modules_15_modules_branch0_modules_bn_buffers_running_mean_"
+    shape = [384]
+    dtype = "torch.float32"
+    device = "cpu"
+    mean = -0.233
+    std = 0.593
+    data = None
+
+
+class Program_weight_tensor_meta_L_self_modules_features_modules_15_modules_branch0_modules_bn_buffers_running_var_:
+    name = "L_self_modules_features_modules_15_modules_branch0_modules_bn_buffers_running_var_"
+    shape = [384]
+    dtype = "torch.float32"
+    device = "cpu"
+    mean = 0.862
+    std = 0.623
+    data = None
+
+
+class Program_weight_tensor_meta_L_self_modules_features_modules_15_modules_branch0_modules_bn_parameters_weight_:
+    name = "L_self_modules_features_modules_15_modules_branch0_modules_bn_parameters_weight_"
+    shape = [384]
+    dtype = "torch.float32"
+    device = "cpu"
+    mean = 1.000
+    std = 0.000
+    data = None
+
+
+class Program_weight_tensor_meta_L_self_modules_features_modules_15_modules_branch0_modules_bn_parameters_bias_:
+    name = (
+        "L_self_modules_features_modules_15_modules_branch0_modules_bn_parameters_bias_"
+    )
+    shape = [384]
+    dtype = "torch.float32"
+    device = "cpu"
+    mean = -0.341
+    std = 0.442
+    data = None
+
+
+class Program_weight_tensor_meta_L_self_modules_features_modules_15_modules_branch1_modules_0_modules_conv_parameters_weight_:
+    name = "L_self_modules_features_modules_15_modules_branch1_modules_0_modules_conv_parameters_weight_"
+    shape = [192, 1024, 1, 1]
+    dtype = "torch.float32"
+    device = "cpu"
+    mean = -0.001
+    std = 0.022
+    data = None
+
+
+class Program_weight_tensor_meta_L_self_modules_features_modules_15_modules_branch1_modules_0_modules_bn_buffers_running_mean_:
+    name = "L_self_modules_features_modules_15_modules_branch1_modules_0_modules_bn_buffers_running_mean_"
+    shape = [192]
+    dtype = "torch.float32"
+    device = "cpu"
+    mean = -0.269
+    std = 0.465
+    data = None
+
+
+class Program_weight_tensor_meta_L_self_modules_features_modules_15_modules_branch1_modules_0_modules_bn_buffers_running_var_:
+    name = "L_self_modules_features_modules_15_modules_branch1_modules_0_modules_bn_buffers_running_var_"
+    shape = [192]
+    dtype = "torch.float32"
+    device = "cpu"
+    mean = 0.511
+    std = 0.307
+    data = None
+
+
+class Program_weight_tensor_meta_L_self_modules_features_modules_15_modules_branch1_modules_0_modules_bn_parameters_weight_:
+    name = "L_self_modules_features_modules_15_modules_branch1_modules_0_modules_bn_parameters_weight_"
+    shape = [192]
+    dtype = "torch.float32"
+    device = "cpu"
+    mean = 1.000
+    std = 0.000
+    data = None
+
+
+class Program_weight_tensor_meta_L_self_modules_features_modules_15_modules_branch1_modules_0_modules_bn_parameters_bias_:
+    name = "L_self_modules_features_modules_15_modules_branch1_modules_0_modules_bn_parameters_bias_"
+    shape = [192]
+    dtype = "torch.float32"
+    device = "cpu"
+    mean = -0.597
+    std = 0.296
+    data = None
+
+
+class Program_weight_tensor_meta_L_self_modules_features_modules_15_modules_branch1_modules_1_modules_conv_parameters_weight_:
+    name = "L_self_modules_features_modules_15_modules_branch1_modules_1_modules_conv_parameters_weight_"
+    shape = [224, 192, 1, 7]
+    dtype = "torch.float32"
+    device = "cpu"
+    mean = -0.001
+    std = 0.019
+    data = None
+
+
+class Program_weight_tensor_meta_L_self_modules_features_modules_15_modules_branch1_modules_1_modules_bn_buffers_running_mean_:
+    name = "L_self_modules_features_modules_15_modules_branch1_modules_1_modules_bn_buffers_running_mean_"
+    shape = [224]
+    dtype = "torch.float32"
+    device = "cpu"
+    mean = -0.299
+    std = 0.276
+    data = None
+
+
+class Program_weight_tensor_meta_L_self_modules_features_modules_15_modules_branch1_modules_1_modules_bn_buffers_running_var_:
+    name = "L_self_modules_features_modules_15_modules_branch1_modules_1_modules_bn_buffers_running_var_"
+    shape = [224]
+    dtype = "torch.float32"
+    device = "cpu"
+    mean = 0.289
+    std = 0.152
+    data = None
+
+
+class Program_weight_tensor_meta_L_self_modules_features_modules_15_modules_branch1_modules_1_modules_bn_parameters_weight_:
+    name = "L_self_modules_features_modules_15_modules_branch1_modules_1_modules_bn_parameters_weight_"
+    shape = [224]
+    dtype = "torch.float32"
+    device = "cpu"
+    mean = 1.000
+    std = 0.000
+    data = None
+
+
+class Program_weight_tensor_meta_L_self_modules_features_modules_15_modules_branch1_modules_1_modules_bn_parameters_bias_:
+    name = "L_self_modules_features_modules_15_modules_branch1_modules_1_modules_bn_parameters_bias_"
+    shape = [224]
+    dtype = "torch.float32"
+    device = "cpu"
+    mean = -0.625
+    std = 0.280
+    data = None
+
+
+class Program_weight_tensor_meta_L_self_modules_features_modules_15_modules_branch1_modules_2_modules_conv_parameters_weight_:
+    name = "L_self_modules_features_modules_15_modules_branch1_modules_2_modules_conv_parameters_weight_"
+    shape = [256, 224, 7, 1]
+    dtype = "torch.float32"
+    device = "cpu"
+    mean = -0.002
+    std = 0.020
+    data = None
+
+
+class Program_weight_tensor_meta_L_self_modules_features_modules_15_modules_branch1_modules_2_modules_bn_buffers_running_mean_:
+    name = "L_self_modules_features_modules_15_modules_branch1_modules_2_modules_bn_buffers_running_mean_"
+    shape = [256]
+    dtype = "torch.float32"
+    device = "cpu"
+    mean = -0.388
+    std = 0.284
+    data = None
+
+
+class Program_weight_tensor_meta_L_self_modules_features_modules_15_modules_branch1_modules_2_modules_bn_buffers_running_var_:
+    name = "L_self_modules_features_modules_15_modules_branch1_modules_2_modules_bn_buffers_running_var_"
+    shape = [256]
+    dtype = "torch.float32"
+    device = "cpu"
+    mean = 0.294
+    std = 0.065
+    data = None
+
+
+class Program_weight_tensor_meta_L_self_modules_features_modules_15_modules_branch1_modules_2_modules_bn_parameters_weight_:
+    name = "L_self_modules_features_modules_15_modules_branch1_modules_2_modules_bn_parameters_weight_"
+    shape = [256]
+    dtype = "torch.float32"
+    device = "cpu"
+    mean = 1.000
+    std = 0.000
+    data = None
+
+
+class Program_weight_tensor_meta_L_self_modules_features_modules_15_modules_branch1_modules_2_modules_bn_parameters_bias_:
+    name = "L_self_modules_features_modules_15_modules_branch1_modules_2_modules_bn_parameters_bias_"
+    shape = [256]
+    dtype = "torch.float32"
+    device = "cpu"
+    mean = -0.748
+    std = 0.206
+    data = None
+
+
+class Program_weight_tensor_meta_L_self_modules_features_modules_15_modules_branch2_modules_0_modules_conv_parameters_weight_:
+    name = "L_self_modules_features_modules_15_modules_branch2_modules_0_modules_conv_parameters_weight_"
+    shape = [192, 1024, 1, 1]
+    dtype = "torch.float32"
+    device = "cpu"
+    mean = -0.000
+    std = 0.019
+    data = None
+
+
+class Program_weight_tensor_meta_L_self_modules_features_modules_15_modules_branch2_modules_0_modules_bn_buffers_running_mean_:
+    name = "L_self_modules_features_modules_15_modules_branch2_modules_0_modules_bn_buffers_running_mean_"
+    shape = [192]
+    dtype = "torch.float32"
+    device = "cpu"
+    mean = -0.083
+    std = 0.410
+    data = None
+
+
+class Program_weight_tensor_meta_L_self_modules_features_modules_15_modules_branch2_modules_0_modules_bn_buffers_running_var_:
+    name = "L_self_modules_features_modules_15_modules_branch2_modules_0_modules_bn_buffers_running_var_"
+    shape = [192]
+    dtype = "torch.float32"
+    device = "cpu"
+    mean = 0.345
+    std = 0.310
+    data = None
+
+
+class Program_weight_tensor_meta_L_self_modules_features_modules_15_modules_branch2_modules_0_modules_bn_parameters_weight_:
+    name = "L_self_modules_features_modules_15_modules_branch2_modules_0_modules_bn_parameters_weight_"
+    shape = [192]
+    dtype = "torch.float32"
+    device = "cpu"
+    mean = 1.000
+    std = 0.000
+    data = None
+
+
+class Program_weight_tensor_meta_L_self_modules_features_modules_15_modules_branch2_modules_0_modules_bn_parameters_bias_:
+    name = "L_self_modules_features_modules_15_modules_branch2_modules_0_modules_bn_parameters_bias_"
+    shape = [192]
+    dtype = "torch.float32"
+    device = "cpu"
+    mean = -0.368
+    std = 0.439
+    data = None
+
+
+class Program_weight_tensor_meta_L_self_modules_features_modules_15_modules_branch2_modules_1_modules_conv_parameters_weight_:
+    name = "L_self_modules_features_modules_15_modules_branch2_modules_1_modules_conv_parameters_weight_"
+    shape = [192, 192, 7, 1]
+    dtype = "torch.float32"
+    device = "cpu"
+    mean = -0.001
+    std = 0.014
+    data = None
+
+
+class Program_weight_tensor_meta_L_self_modules_features_modules_15_modules_branch2_modules_1_modules_bn_buffers_running_mean_:
+    name = "L_self_modules_features_modules_15_modules_branch2_modules_1_modules_bn_buffers_running_mean_"
+    shape = [192]
+    dtype = "torch.float32"
+    device = "cpu"
+    mean = -0.202
+    std = 0.429
+    data = None
+
+
+class Program_weight_tensor_meta_L_self_modules_features_modules_15_modules_branch2_modules_1_modules_bn_buffers_running_var_:
+    name = "L_self_modules_features_modules_15_modules_branch2_modules_1_modules_bn_buffers_running_var_"
+    shape = [192]
+    dtype = "torch.float32"
+    device = "cpu"
+    mean = 0.328
+    std = 0.248
+    data = None
+
+
+class Program_weight_tensor_meta_L_self_modules_features_modules_15_modules_branch2_modules_1_modules_bn_parameters_weight_:
+    name = "L_self_modules_features_modules_15_modules_branch2_modules_1_modules_bn_parameters_weight_"
+    shape = [192]
+    dtype = "torch.float32"
+    device = "cpu"
+    mean = 1.000
+    std = 0.000
+    data = None
+
+
+class Program_weight_tensor_meta_L_self_modules_features_modules_15_modules_branch2_modules_1_modules_bn_parameters_bias_:
+    name = "L_self_modules_features_modules_15_modules_branch2_modules_1_modules_bn_parameters_bias_"
+    shape = [192]
+    dtype = "torch.float32"
+    device = "cpu"
+    mean = -0.289
+    std = 0.427
+    data = None
+
+
+class Program_weight_tensor_meta_L_self_modules_features_modules_15_modules_branch2_modules_2_modules_conv_parameters_weight_:
+    name = "L_self_modules_features_modules_15_modules_branch2_modules_2_modules_conv_parameters_weight_"
+    shape = [224, 192, 1, 7]
+    dtype = "torch.float32"
+    device = "cpu"
+    mean = -0.001
+    std = 0.014
+    data = None
+
+
+class Program_weight_tensor_meta_L_self_modules_features_modules_15_modules_branch2_modules_2_modules_bn_buffers_running_mean_:
+    name = "L_self_modules_features_modules_15_modules_branch2_modules_2_modules_bn_buffers_running_mean_"
+    shape = [224]
+    dtype = "torch.float32"
+    device = "cpu"
+    mean = -0.294
+    std = 0.472
+    data = None
+
+
+class Program_weight_tensor_meta_L_self_modules_features_modules_15_modules_branch2_modules_2_modules_bn_buffers_running_var_:
+    name = "L_self_modules_features_modules_15_modules_branch2_modules_2_modules_bn_buffers_running_var_"
+    shape = [224]
+    dtype = "torch.float32"
+    device = "cpu"
+    mean = 0.392
+    std = 0.231
+    data = None
+
+
+class Program_weight_tensor_meta_L_self_modules_features_modules_15_modules_branch2_modules_2_modules_bn_parameters_weight_:
+    name = "L_self_modules_features_modules_15_modules_branch2_modules_2_modules_bn_parameters_weight_"
+    shape = [224]
+    dtype = "torch.float32"
+    device = "cpu"
+    mean = 1.000
+    std = 0.000
+    data = None
+
+
+class Program_weight_tensor_meta_L_self_modules_features_modules_15_modules_branch2_modules_2_modules_bn_parameters_bias_:
+    name = "L_self_modules_features_modules_15_modules_branch2_modules_2_modules_bn_parameters_bias_"
+    shape = [224]
+    dtype = "torch.float32"
+    device = "cpu"
+    mean = -0.432
+    std = 0.264
+    data = None
+
+
+class Program_weight_tensor_meta_L_self_modules_features_modules_15_modules_branch2_modules_3_modules_conv_parameters_weight_:
+    name = "L_self_modules_features_modules_15_modules_branch2_modules_3_modules_conv_parameters_weight_"
+    shape = [224, 224, 7, 1]
+    dtype = "torch.float32"
+    device = "cpu"
+    mean = -0.001
+    std = 0.014
+    data = None
+
+
+class Program_weight_tensor_meta_L_self_modules_features_modules_15_modules_branch2_modules_3_modules_bn_buffers_running_mean_:
+    name = "L_self_modules_features_modules_15_modules_branch2_modules_3_modules_bn_buffers_running_mean_"
+    shape = [224]
+    dtype = "torch.float32"
+    device = "cpu"
+    mean = -0.311
+    std = 0.424
+    data = None
+
+
+class Program_weight_tensor_meta_L_self_modules_features_modules_15_modules_branch2_modules_3_modules_bn_buffers_running_var_:
+    name = "L_self_modules_features_modules_15_modules_branch2_modules_3_modules_bn_buffers_running_var_"
+    shape = [224]
+    dtype = "torch.float32"
+    device = "cpu"
+    mean = 0.376
+    std = 0.237
+    data = None
+
+
+class Program_weight_tensor_meta_L_self_modules_features_modules_15_modules_branch2_modules_3_modules_bn_parameters_weight_:
+    name = "L_self_modules_features_modules_15_modules_branch2_modules_3_modules_bn_parameters_weight_"
+    shape = [224]
+    dtype = "torch.float32"
+    device = "cpu"
+    mean = 1.000
+    std = 0.000
+    data = None
+
+
+class Program_weight_tensor_meta_L_self_modules_features_modules_15_modules_branch2_modules_3_modules_bn_parameters_bias_:
+    name = "L_self_modules_features_modules_15_modules_branch2_modules_3_modules_bn_parameters_bias_"
+    shape = [224]
+    dtype = "torch.float32"
+    device = "cpu"
+    mean = -0.444
+    std = 0.333
+    data = None
+
+
+class Program_weight_tensor_meta_L_self_modules_features_modules_15_modules_branch2_modules_4_modules_conv_parameters_weight_:
+    name = "L_self_modules_features_modules_15_modules_branch2_modules_4_modules_conv_parameters_weight_"
+    shape = [256, 224, 1, 7]
+    dtype = "torch.float32"
+    device = "cpu"
+    mean = -0.001
+    std = 0.014
+    data = None
+
+
+class Program_weight_tensor_meta_L_self_modules_features_modules_15_modules_branch2_modules_4_modules_bn_buffers_running_mean_:
+    name = "L_self_modules_features_modules_15_modules_branch2_modules_4_modules_bn_buffers_running_mean_"
+    shape = [256]
+    dtype = "torch.float32"
+    device = "cpu"
+    mean = -0.551
+    std = 0.428
+    data = None
+
+
+class Program_weight_tensor_meta_L_self_modules_features_modules_15_modules_branch2_modules_4_modules_bn_buffers_running_var_:
+    name = "L_self_modules_features_modules_15_modules_branch2_modules_4_modules_bn_buffers_running_var_"
+    shape = [256]
+    dtype = "torch.float32"
+    device = "cpu"
+    mean = 0.409
+    std = 0.295
+    data = None
+
+
+class Program_weight_tensor_meta_L_self_modules_features_modules_15_modules_branch2_modules_4_modules_bn_parameters_weight_:
+    name = "L_self_modules_features_modules_15_modules_branch2_modules_4_modules_bn_parameters_weight_"
+    shape = [256]
+    dtype = "torch.float32"
+    device = "cpu"
+    mean = 1.000
+    std = 0.000
+    data = None
+
+
+class Program_weight_tensor_meta_L_self_modules_features_modules_15_modules_branch2_modules_4_modules_bn_parameters_bias_:
+    name = "L_self_modules_features_modules_15_modules_branch2_modules_4_modules_bn_parameters_bias_"
+    shape = [256]
+    dtype = "torch.float32"
+    device = "cpu"
+    mean = -0.800
+    std = 0.275
+    data = None
+
+
+class Program_weight_tensor_meta_L_self_modules_features_modules_15_modules_branch3_modules_1_modules_conv_parameters_weight_:
+    name = "L_self_modules_features_modules_15_modules_branch3_modules_1_modules_conv_parameters_weight_"
+    shape = [128, 1024, 1, 1]
+    dtype = "torch.float32"
+    device = "cpu"
+    mean = -0.001
+    std = 0.028
+    data = None
+
+
+class Program_weight_tensor_meta_L_self_modules_features_modules_15_modules_branch3_modules_1_modules_bn_buffers_running_mean_:
+    name = "L_self_modules_features_modules_15_modules_branch3_modules_1_modules_bn_buffers_running_mean_"
+    shape = [128]
+    dtype = "torch.float32"
+    device = "cpu"
+    mean = -0.163
+    std = 0.371
+    data = None
+
+
+class Program_weight_tensor_meta_L_self_modules_features_modules_15_modules_branch3_modules_1_modules_bn_buffers_running_var_:
+    name = "L_self_modules_features_modules_15_modules_branch3_modules_1_modules_bn_buffers_running_var_"
+    shape = [128]
+    dtype = "torch.float32"
+    device = "cpu"
+    mean = 0.307
+    std = 0.103
+    data = None
+
+
+class Program_weight_tensor_meta_L_self_modules_features_modules_15_modules_branch3_modules_1_modules_bn_parameters_weight_:
+    name = "L_self_modules_features_modules_15_modules_branch3_modules_1_modules_bn_parameters_weight_"
+    shape = [128]
+    dtype = "torch.float32"
+    device = "cpu"
+    mean = 1.000
+    std = 0.000
+    data = None
+
+
+class Program_weight_tensor_meta_L_self_modules_features_modules_15_modules_branch3_modules_1_modules_bn_parameters_bias_:
+    name = "L_self_modules_features_modules_15_modules_branch3_modules_1_modules_bn_parameters_bias_"
+    shape = [128]
+    dtype = "torch.float32"
+    device = "cpu"
+    mean = -0.667
+    std = 0.172
+    data = None
+
+
+class Program_weight_tensor_meta_L_self_modules_features_modules_16_modules_branch0_modules_conv_parameters_weight_:
+    name = "L_self_modules_features_modules_16_modules_branch0_modules_conv_parameters_weight_"
+    shape = [384, 1024, 1, 1]
+    dtype = "torch.float32"
+    device = "cpu"
+    mean = -0.002
+    std = 0.032
+    data = None
+
+
+class Program_weight_tensor_meta_L_self_modules_features_modules_16_modules_branch0_modules_bn_buffers_running_mean_:
+    name = "L_self_modules_features_modules_16_modules_branch0_modules_bn_buffers_running_mean_"
+    shape = [384]
+    dtype = "torch.float32"
+    device = "cpu"
+    mean = -0.342
+    std = 0.287
+    data = None
+
+
+class Program_weight_tensor_meta_L_self_modules_features_modules_16_modules_branch0_modules_bn_buffers_running_var_:
+    name = "L_self_modules_features_modules_16_modules_branch0_modules_bn_buffers_running_var_"
+    shape = [384]
+    dtype = "torch.float32"
+    device = "cpu"
+    mean = 0.486
+    std = 0.313
+    data = None
+
+
+class Program_weight_tensor_meta_L_self_modules_features_modules_16_modules_branch0_modules_bn_parameters_weight_:
+    name = "L_self_modules_features_modules_16_modules_branch0_modules_bn_parameters_weight_"
+    shape = [384]
+    dtype = "torch.float32"
+    device = "cpu"
+    mean = 1.000
+    std = 0.000
+    data = None
+
+
+class Program_weight_tensor_meta_L_self_modules_features_modules_16_modules_branch0_modules_bn_parameters_bias_:
+    name = (
+        "L_self_modules_features_modules_16_modules_branch0_modules_bn_parameters_bias_"
+    )
+    shape = [384]
+    dtype = "torch.float32"
+    device = "cpu"
+    mean = -0.705
+    std = 0.363
+    data = None
+
+
+class Program_weight_tensor_meta_L_self_modules_features_modules_16_modules_branch1_modules_0_modules_conv_parameters_weight_:
+    name = "L_self_modules_features_modules_16_modules_branch1_modules_0_modules_conv_parameters_weight_"
+    shape = [192, 1024, 1, 1]
+    dtype = "torch.float32"
+    device = "cpu"
+    mean = -0.001
+    std = 0.025
+    data = None
+
+
+class Program_weight_tensor_meta_L_self_modules_features_modules_16_modules_branch1_modules_0_modules_bn_buffers_running_mean_:
+    name = "L_self_modules_features_modules_16_modules_branch1_modules_0_modules_bn_buffers_running_mean_"
+    shape = [192]
+    dtype = "torch.float32"
+    device = "cpu"
+    mean = -0.296
+    std = 0.249
+    data = None
+
+
+class Program_weight_tensor_meta_L_self_modules_features_modules_16_modules_branch1_modules_0_modules_bn_buffers_running_var_:
+    name = "L_self_modules_features_modules_16_modules_branch1_modules_0_modules_bn_buffers_running_var_"
+    shape = [192]
+    dtype = "torch.float32"
+    device = "cpu"
+    mean = 0.336
+    std = 0.180
+    data = None
+
+
+class Program_weight_tensor_meta_L_self_modules_features_modules_16_modules_branch1_modules_0_modules_bn_parameters_weight_:
+    name = "L_self_modules_features_modules_16_modules_branch1_modules_0_modules_bn_parameters_weight_"
+    shape = [192]
+    dtype = "torch.float32"
+    device = "cpu"
+    mean = 1.000
+    std = 0.000
+    data = None
+
+
+class Program_weight_tensor_meta_L_self_modules_features_modules_16_modules_branch1_modules_0_modules_bn_parameters_bias_:
+    name = "L_self_modules_features_modules_16_modules_branch1_modules_0_modules_bn_parameters_bias_"
+    shape = [192]
+    dtype = "torch.float32"
+    device = "cpu"
+    mean = -0.713
+    std = 0.307
+    data = None
+
+
+class Program_weight_tensor_meta_L_self_modules_features_modules_16_modules_branch1_modules_1_modules_conv_parameters_weight_:
+    name = "L_self_modules_features_modules_16_modules_branch1_modules_1_modules_conv_parameters_weight_"
+    shape = [224, 192, 1, 7]
+    dtype = "torch.float32"
+    device = "cpu"
+    mean = -0.001
+    std = 0.020
+    data = None
+
+
+class Program_weight_tensor_meta_L_self_modules_features_modules_16_modules_branch1_modules_1_modules_bn_buffers_running_mean_:
+    name = "L_self_modules_features_modules_16_modules_branch1_modules_1_modules_bn_buffers_running_mean_"
+    shape = [224]
+    dtype = "torch.float32"
+    device = "cpu"
+    mean = -0.221
+    std = 0.227
+    data = None
+
+
+class Program_weight_tensor_meta_L_self_modules_features_modules_16_modules_branch1_modules_1_modules_bn_buffers_running_var_:
+    name = "L_self_modules_features_modules_16_modules_branch1_modules_1_modules_bn_buffers_running_var_"
+    shape = [224]
+    dtype = "torch.float32"
+    device = "cpu"
+    mean = 0.254
+    std = 0.195
+    data = None
+
+
+class Program_weight_tensor_meta_L_self_modules_features_modules_16_modules_branch1_modules_1_modules_bn_parameters_weight_:
+    name = "L_self_modules_features_modules_16_modules_branch1_modules_1_modules_bn_parameters_weight_"
+    shape = [224]
+    dtype = "torch.float32"
+    device = "cpu"
+    mean = 1.000
+    std = 0.000
+    data = None
+
+
+class Program_weight_tensor_meta_L_self_modules_features_modules_16_modules_branch1_modules_1_modules_bn_parameters_bias_:
+    name = "L_self_modules_features_modules_16_modules_branch1_modules_1_modules_bn_parameters_bias_"
+    shape = [224]
+    dtype = "torch.float32"
+    device = "cpu"
+    mean = -0.647
+    std = 0.346
+    data = None
+
+
+class Program_weight_tensor_meta_L_self_modules_features_modules_16_modules_branch1_modules_2_modules_conv_parameters_weight_:
+    name = "L_self_modules_features_modules_16_modules_branch1_modules_2_modules_conv_parameters_weight_"
+    shape = [256, 224, 7, 1]
+    dtype = "torch.float32"
+    device = "cpu"
+    mean = -0.002
+    std = 0.021
+    data = None
+
+
+class Program_weight_tensor_meta_L_self_modules_features_modules_16_modules_branch1_modules_2_modules_bn_buffers_running_mean_:
+    name = "L_self_modules_features_modules_16_modules_branch1_modules_2_modules_bn_buffers_running_mean_"
+    shape = [256]
+    dtype = "torch.float32"
+    device = "cpu"
+    mean = -0.411
+    std = 0.334
+    data = None
+
+
+class Program_weight_tensor_meta_L_self_modules_features_modules_16_modules_branch1_modules_2_modules_bn_buffers_running_var_:
+    name = "L_self_modules_features_modules_16_modules_branch1_modules_2_modules_bn_buffers_running_var_"
+    shape = [256]
+    dtype = "torch.float32"
+    device = "cpu"
+    mean = 0.310
+    std = 0.141
+    data = None
+
+
+class Program_weight_tensor_meta_L_self_modules_features_modules_16_modules_branch1_modules_2_modules_bn_parameters_weight_:
+    name = "L_self_modules_features_modules_16_modules_branch1_modules_2_modules_bn_parameters_weight_"
+    shape = [256]
+    dtype = "torch.float32"
+    device = "cpu"
+    mean = 1.000
+    std = 0.000
+    data = None
+
+
+class Program_weight_tensor_meta_L_self_modules_features_modules_16_modules_branch1_modules_2_modules_bn_parameters_bias_:
+    name = "L_self_modules_features_modules_16_modules_branch1_modules_2_modules_bn_parameters_bias_"
+    shape = [256]
+    dtype = "torch.float32"
+    device = "cpu"
+    mean = -0.961
+    std = 0.298
+    data = None
+
+
+class Program_weight_tensor_meta_L_self_modules_features_modules_16_modules_branch2_modules_0_modules_conv_parameters_weight_:
+    name = "L_self_modules_features_modules_16_modules_branch2_modules_0_modules_conv_parameters_weight_"
+    shape = [192, 1024, 1, 1]
+    dtype = "torch.float32"
+    device = "cpu"
+    mean = -0.001
+    std = 0.026
+    data = None
+
+
+class Program_weight_tensor_meta_L_self_modules_features_modules_16_modules_branch2_modules_0_modules_bn_buffers_running_mean_:
+    name = "L_self_modules_features_modules_16_modules_branch2_modules_0_modules_bn_buffers_running_mean_"
+    shape = [192]
+    dtype = "torch.float32"
+    device = "cpu"
+    mean = -0.243
+    std = 0.304
+    data = None
+
+
+class Program_weight_tensor_meta_L_self_modules_features_modules_16_modules_branch2_modules_0_modules_bn_buffers_running_var_:
+    name = "L_self_modules_features_modules_16_modules_branch2_modules_0_modules_bn_buffers_running_var_"
+    shape = [192]
+    dtype = "torch.float32"
+    device = "cpu"
+    mean = 0.381
+    std = 0.275
+    data = None
+
+
+class Program_weight_tensor_meta_L_self_modules_features_modules_16_modules_branch2_modules_0_modules_bn_parameters_weight_:
+    name = "L_self_modules_features_modules_16_modules_branch2_modules_0_modules_bn_parameters_weight_"
+    shape = [192]
+    dtype = "torch.float32"
+    device = "cpu"
+    mean = 1.000
+    std = 0.000
+    data = None
+
+
+class Program_weight_tensor_meta_L_self_modules_features_modules_16_modules_branch2_modules_0_modules_bn_parameters_bias_:
+    name = "L_self_modules_features_modules_16_modules_branch2_modules_0_modules_bn_parameters_bias_"
+    shape = [192]
+    dtype = "torch.float32"
+    device = "cpu"
+    mean = -0.686
+    std = 0.377
+    data = None
+
+
+class Program_weight_tensor_meta_L_self_modules_features_modules_16_modules_branch2_modules_1_modules_conv_parameters_weight_:
+    name = "L_self_modules_features_modules_16_modules_branch2_modules_1_modules_conv_parameters_weight_"
+    shape = [192, 192, 7, 1]
+    dtype = "torch.float32"
+    device = "cpu"
+    mean = -0.002
+    std = 0.023
+    data = None
+
+
+class Program_weight_tensor_meta_L_self_modules_features_modules_16_modules_branch2_modules_1_modules_bn_buffers_running_mean_:
+    name = "L_self_modules_features_modules_16_modules_branch2_modules_1_modules_bn_buffers_running_mean_"
+    shape = [192]
+    dtype = "torch.float32"
+    device = "cpu"
+    mean = -0.349
+    std = 0.236
+    data = None
+
+
+class Program_weight_tensor_meta_L_self_modules_features_modules_16_modules_branch2_modules_1_modules_bn_buffers_running_var_:
+    name = "L_self_modules_features_modules_16_modules_branch2_modules_1_modules_bn_buffers_running_var_"
+    shape = [192]
+    dtype = "torch.float32"
+    device = "cpu"
+    mean = 0.317
+    std = 0.130
+    data = None
+
+
+class Program_weight_tensor_meta_L_self_modules_features_modules_16_modules_branch2_modules_1_modules_bn_parameters_weight_:
+    name = "L_self_modules_features_modules_16_modules_branch2_modules_1_modules_bn_parameters_weight_"
+    shape = [192]
+    dtype = "torch.float32"
+    device = "cpu"
+    mean = 1.000
+    std = 0.000
+    data = None
+
+
+class Program_weight_tensor_meta_L_self_modules_features_modules_16_modules_branch2_modules_1_modules_bn_parameters_bias_:
+    name = "L_self_modules_features_modules_16_modules_branch2_modules_1_modules_bn_parameters_bias_"
+    shape = [192]
+    dtype = "torch.float32"
+    device = "cpu"
+    mean = -0.519
+    std = 0.326
+    data = None
+
+
+class Program_weight_tensor_meta_L_self_modules_features_modules_16_modules_branch2_modules_2_modules_conv_parameters_weight_:
+    name = "L_self_modules_features_modules_16_modules_branch2_modules_2_modules_conv_parameters_weight_"
+    shape = [224, 192, 1, 7]
+    dtype = "torch.float32"
+    device = "cpu"
+    mean = -0.002
+    std = 0.022
+    data = None
+
+
+class Program_weight_tensor_meta_L_self_modules_features_modules_16_modules_branch2_modules_2_modules_bn_buffers_running_mean_:
+    name = "L_self_modules_features_modules_16_modules_branch2_modules_2_modules_bn_buffers_running_mean_"
+    shape = [224]
+    dtype = "torch.float32"
+    device = "cpu"
+    mean = -0.461
+    std = 0.287
+    data = None
+
+
+class Program_weight_tensor_meta_L_self_modules_features_modules_16_modules_branch2_modules_2_modules_bn_buffers_running_var_:
+    name = "L_self_modules_features_modules_16_modules_branch2_modules_2_modules_bn_buffers_running_var_"
+    shape = [224]
+    dtype = "torch.float32"
+    device = "cpu"
+    mean = 0.397
+    std = 0.157
+    data = None
+
+
+class Program_weight_tensor_meta_L_self_modules_features_modules_16_modules_branch2_modules_2_modules_bn_parameters_weight_:
+    name = "L_self_modules_features_modules_16_modules_branch2_modules_2_modules_bn_parameters_weight_"
+    shape = [224]
+    dtype = "torch.float32"
+    device = "cpu"
+    mean = 1.000
+    std = 0.000
+    data = None
+
+
+class Program_weight_tensor_meta_L_self_modules_features_modules_16_modules_branch2_modules_2_modules_bn_parameters_bias_:
+    name = "L_self_modules_features_modules_16_modules_branch2_modules_2_modules_bn_parameters_bias_"
+    shape = [224]
+    dtype = "torch.float32"
+    device = "cpu"
+    mean = -0.920
+    std = 0.262
+    data = None
+
+
+class Program_weight_tensor_meta_L_self_modules_features_modules_16_modules_branch2_modules_3_modules_conv_parameters_weight_:
+    name = "L_self_modules_features_modules_16_modules_branch2_modules_3_modules_conv_parameters_weight_"
+    shape = [224, 224, 7, 1]
+    dtype = "torch.float32"
+    device = "cpu"
+    mean = -0.002
+    std = 0.022
+    data = None
+
+
+class Program_weight_tensor_meta_L_self_modules_features_modules_16_modules_branch2_modules_3_modules_bn_buffers_running_mean_:
+    name = "L_self_modules_features_modules_16_modules_branch2_modules_3_modules_bn_buffers_running_mean_"
+    shape = [224]
+    dtype = "torch.float32"
+    device = "cpu"
+    mean = -0.307
+    std = 0.162
+    data = None
+
+
+class Program_weight_tensor_meta_L_self_modules_features_modules_16_modules_branch2_modules_3_modules_bn_buffers_running_var_:
+    name = "L_self_modules_features_modules_16_modules_branch2_modules_3_modules_bn_buffers_running_var_"
+    shape = [224]
+    dtype = "torch.float32"
+    device = "cpu"
+    mean = 0.246
+    std = 0.066
+    data = None
+
+
+class Program_weight_tensor_meta_L_self_modules_features_modules_16_modules_branch2_modules_3_modules_bn_parameters_weight_:
+    name = "L_self_modules_features_modules_16_modules_branch2_modules_3_modules_bn_parameters_weight_"
+    shape = [224]
+    dtype = "torch.float32"
+    device = "cpu"
+    mean = 1.000
+    std = 0.000
+    data = None
+
+
+class Program_weight_tensor_meta_L_self_modules_features_modules_16_modules_branch2_modules_3_modules_bn_parameters_bias_:
+    name = "L_self_modules_features_modules_16_modules_branch2_modules_3_modules_bn_parameters_bias_"
+    shape = [224]
+    dtype = "torch.float32"
+    device = "cpu"
+    mean = -0.582
+    std = 0.222
+    data = None
+
+
+class Program_weight_tensor_meta_L_self_modules_features_modules_16_modules_branch2_modules_4_modules_conv_parameters_weight_:
+    name = "L_self_modules_features_modules_16_modules_branch2_modules_4_modules_conv_parameters_weight_"
+    shape = [256, 224, 1, 7]
+    dtype = "torch.float32"
+    device = "cpu"
+    mean = -0.003
+    std = 0.022
+    data = None
+
+
+class Program_weight_tensor_meta_L_self_modules_features_modules_16_modules_branch2_modules_4_modules_bn_buffers_running_mean_:
+    name = "L_self_modules_features_modules_16_modules_branch2_modules_4_modules_bn_buffers_running_mean_"
+    shape = [256]
+    dtype = "torch.float32"
+    device = "cpu"
+    mean = -0.703
+    std = 0.315
+    data = None
+
+
+class Program_weight_tensor_meta_L_self_modules_features_modules_16_modules_branch2_modules_4_modules_bn_buffers_running_var_:
+    name = "L_self_modules_features_modules_16_modules_branch2_modules_4_modules_bn_buffers_running_var_"
+    shape = [256]
+    dtype = "torch.float32"
+    device = "cpu"
+    mean = 0.484
+    std = 0.240
+    data = None
+
+
+class Program_weight_tensor_meta_L_self_modules_features_modules_16_modules_branch2_modules_4_modules_bn_parameters_weight_:
+    name = "L_self_modules_features_modules_16_modules_branch2_modules_4_modules_bn_parameters_weight_"
+    shape = [256]
+    dtype = "torch.float32"
+    device = "cpu"
+    mean = 1.000
+    std = 0.000
+    data = None
+
+
+class Program_weight_tensor_meta_L_self_modules_features_modules_16_modules_branch2_modules_4_modules_bn_parameters_bias_:
+    name = "L_self_modules_features_modules_16_modules_branch2_modules_4_modules_bn_parameters_bias_"
+    shape = [256]
+    dtype = "torch.float32"
+    device = "cpu"
+    mean = -0.789
+    std = 0.303
+    data = None
+
+
+class Program_weight_tensor_meta_L_self_modules_features_modules_16_modules_branch3_modules_1_modules_conv_parameters_weight_:
+    name = "L_self_modules_features_modules_16_modules_branch3_modules_1_modules_conv_parameters_weight_"
+    shape = [128, 1024, 1, 1]
+    dtype = "torch.float32"
+    device = "cpu"
+    mean = -0.002
+    std = 0.031
+    data = None
+
+
+class Program_weight_tensor_meta_L_self_modules_features_modules_16_modules_branch3_modules_1_modules_bn_buffers_running_mean_:
+    name = "L_self_modules_features_modules_16_modules_branch3_modules_1_modules_bn_buffers_running_mean_"
+    shape = [128]
+    dtype = "torch.float32"
+    device = "cpu"
+    mean = -0.380
+    std = 0.207
+    data = None
+
+
+class Program_weight_tensor_meta_L_self_modules_features_modules_16_modules_branch3_modules_1_modules_bn_buffers_running_var_:
+    name = "L_self_modules_features_modules_16_modules_branch3_modules_1_modules_bn_buffers_running_var_"
+    shape = [128]
+    dtype = "torch.float32"
+    device = "cpu"
+    mean = 0.198
+    std = 0.139
+    data = None
+
+
+class Program_weight_tensor_meta_L_self_modules_features_modules_16_modules_branch3_modules_1_modules_bn_parameters_weight_:
+    name = "L_self_modules_features_modules_16_modules_branch3_modules_1_modules_bn_parameters_weight_"
+    shape = [128]
+    dtype = "torch.float32"
+    device = "cpu"
+    mean = 1.000
+    std = 0.000
+    data = None
+
+
+class Program_weight_tensor_meta_L_self_modules_features_modules_16_modules_branch3_modules_1_modules_bn_parameters_bias_:
+    name = "L_self_modules_features_modules_16_modules_branch3_modules_1_modules_bn_parameters_bias_"
+    shape = [128]
+    dtype = "torch.float32"
+    device = "cpu"
+    mean = -0.953
+    std = 0.274
+    data = None
+
+
+class Program_weight_tensor_meta_L_self_modules_features_modules_17_modules_branch0_modules_conv_parameters_weight_:
+    name = "L_self_modules_features_modules_17_modules_branch0_modules_conv_parameters_weight_"
+    shape = [384, 1024, 1, 1]
+    dtype = "torch.float32"
+    device = "cpu"
+    mean = -0.003
+    std = 0.034
+    data = None
+
+
+class Program_weight_tensor_meta_L_self_modules_features_modules_17_modules_branch0_modules_bn_buffers_running_mean_:
+    name = "L_self_modules_features_modules_17_modules_branch0_modules_bn_buffers_running_mean_"
+    shape = [384]
+    dtype = "torch.float32"
+    device = "cpu"
+    mean = -0.412
+    std = 0.175
+    data = None
+
+
+class Program_weight_tensor_meta_L_self_modules_features_modules_17_modules_branch0_modules_bn_buffers_running_var_:
+    name = "L_self_modules_features_modules_17_modules_branch0_modules_bn_buffers_running_var_"
+    shape = [384]
+    dtype = "torch.float32"
+    device = "cpu"
+    mean = 0.341
+    std = 0.115
+    data = None
+
+
+class Program_weight_tensor_meta_L_self_modules_features_modules_17_modules_branch0_modules_bn_parameters_weight_:
+    name = "L_self_modules_features_modules_17_modules_branch0_modules_bn_parameters_weight_"
+    shape = [384]
+    dtype = "torch.float32"
+    device = "cpu"
+    mean = 1.000
+    std = 0.000
+    data = None
+
+
+class Program_weight_tensor_meta_L_self_modules_features_modules_17_modules_branch0_modules_bn_parameters_bias_:
+    name = (
+        "L_self_modules_features_modules_17_modules_branch0_modules_bn_parameters_bias_"
+    )
+    shape = [384]
+    dtype = "torch.float32"
+    device = "cpu"
+    mean = -1.172
+    std = 0.227
+    data = None
+
+
+class Program_weight_tensor_meta_L_self_modules_features_modules_17_modules_branch1_modules_0_modules_conv_parameters_weight_:
+    name = "L_self_modules_features_modules_17_modules_branch1_modules_0_modules_conv_parameters_weight_"
+    shape = [192, 1024, 1, 1]
+    dtype = "torch.float32"
+    device = "cpu"
+    mean = -0.003
+    std = 0.035
+    data = None
+
+
+class Program_weight_tensor_meta_L_self_modules_features_modules_17_modules_branch1_modules_0_modules_bn_buffers_running_mean_:
+    name = "L_self_modules_features_modules_17_modules_branch1_modules_0_modules_bn_buffers_running_mean_"
+    shape = [192]
+    dtype = "torch.float32"
+    device = "cpu"
+    mean = -0.290
+    std = 0.196
+    data = None
+
+
+class Program_weight_tensor_meta_L_self_modules_features_modules_17_modules_branch1_modules_0_modules_bn_buffers_running_var_:
+    name = "L_self_modules_features_modules_17_modules_branch1_modules_0_modules_bn_buffers_running_var_"
+    shape = [192]
+    dtype = "torch.float32"
+    device = "cpu"
+    mean = 0.310
+    std = 0.138
+    data = None
+
+
+class Program_weight_tensor_meta_L_self_modules_features_modules_17_modules_branch1_modules_0_modules_bn_parameters_weight_:
+    name = "L_self_modules_features_modules_17_modules_branch1_modules_0_modules_bn_parameters_weight_"
+    shape = [192]
+    dtype = "torch.float32"
+    device = "cpu"
+    mean = 1.000
+    std = 0.000
+    data = None
+
+
+class Program_weight_tensor_meta_L_self_modules_features_modules_17_modules_branch1_modules_0_modules_bn_parameters_bias_:
+    name = "L_self_modules_features_modules_17_modules_branch1_modules_0_modules_bn_parameters_bias_"
+    shape = [192]
+    dtype = "torch.float32"
+    device = "cpu"
+    mean = -0.911
+    std = 0.326
+    data = None
+
+
+class Program_weight_tensor_meta_L_self_modules_features_modules_17_modules_branch1_modules_1_modules_conv_parameters_weight_:
+    name = "L_self_modules_features_modules_17_modules_branch1_modules_1_modules_conv_parameters_weight_"
+    shape = [224, 192, 1, 7]
+    dtype = "torch.float32"
+    device = "cpu"
+    mean = -0.002
+    std = 0.025
+    data = None
+
+
+class Program_weight_tensor_meta_L_self_modules_features_modules_17_modules_branch1_modules_1_modules_bn_buffers_running_mean_:
+    name = "L_self_modules_features_modules_17_modules_branch1_modules_1_modules_bn_buffers_running_mean_"
+    shape = [224]
+    dtype = "torch.float32"
+    device = "cpu"
+    mean = -0.225
+    std = 0.133
+    data = None
+
+
+class Program_weight_tensor_meta_L_self_modules_features_modules_17_modules_branch1_modules_1_modules_bn_buffers_running_var_:
+    name = "L_self_modules_features_modules_17_modules_branch1_modules_1_modules_bn_buffers_running_var_"
+    shape = [224]
+    dtype = "torch.float32"
+    device = "cpu"
+    mean = 0.204
+    std = 0.085
+    data = None
+
+
+class Program_weight_tensor_meta_L_self_modules_features_modules_17_modules_branch1_modules_1_modules_bn_parameters_weight_:
+    name = "L_self_modules_features_modules_17_modules_branch1_modules_1_modules_bn_parameters_weight_"
+    shape = [224]
+    dtype = "torch.float32"
+    device = "cpu"
+    mean = 1.000
+    std = 0.000
+    data = None
+
+
+class Program_weight_tensor_meta_L_self_modules_features_modules_17_modules_branch1_modules_1_modules_bn_parameters_bias_:
+    name = "L_self_modules_features_modules_17_modules_branch1_modules_1_modules_bn_parameters_bias_"
+    shape = [224]
+    dtype = "torch.float32"
+    device = "cpu"
+    mean = -0.549
+    std = 0.309
+    data = None
+
+
+class Program_weight_tensor_meta_L_self_modules_features_modules_17_modules_branch1_modules_2_modules_conv_parameters_weight_:
+    name = "L_self_modules_features_modules_17_modules_branch1_modules_2_modules_conv_parameters_weight_"
+    shape = [256, 224, 7, 1]
+    dtype = "torch.float32"
+    device = "cpu"
+    mean = -0.003
+    std = 0.024
+    data = None
+
+
+class Program_weight_tensor_meta_L_self_modules_features_modules_17_modules_branch1_modules_2_modules_bn_buffers_running_mean_:
+    name = "L_self_modules_features_modules_17_modules_branch1_modules_2_modules_bn_buffers_running_mean_"
+    shape = [256]
+    dtype = "torch.float32"
+    device = "cpu"
+    mean = -0.813
+    std = 0.199
+    data = None
+
+
+class Program_weight_tensor_meta_L_self_modules_features_modules_17_modules_branch1_modules_2_modules_bn_buffers_running_var_:
+    name = "L_self_modules_features_modules_17_modules_branch1_modules_2_modules_bn_buffers_running_var_"
+    shape = [256]
+    dtype = "torch.float32"
+    device = "cpu"
+    mean = 0.381
+    std = 0.093
+    data = None
+
+
+class Program_weight_tensor_meta_L_self_modules_features_modules_17_modules_branch1_modules_2_modules_bn_parameters_weight_:
+    name = "L_self_modules_features_modules_17_modules_branch1_modules_2_modules_bn_parameters_weight_"
+    shape = [256]
+    dtype = "torch.float32"
+    device = "cpu"
+    mean = 1.000
+    std = 0.000
+    data = None
+
+
+class Program_weight_tensor_meta_L_self_modules_features_modules_17_modules_branch1_modules_2_modules_bn_parameters_bias_:
+    name = "L_self_modules_features_modules_17_modules_branch1_modules_2_modules_bn_parameters_bias_"
+    shape = [256]
+    dtype = "torch.float32"
+    device = "cpu"
+    mean = -1.560
+    std = 0.254
+    data = None
+
+
+class Program_weight_tensor_meta_L_self_modules_features_modules_17_modules_branch2_modules_0_modules_conv_parameters_weight_:
+    name = "L_self_modules_features_modules_17_modules_branch2_modules_0_modules_conv_parameters_weight_"
+    shape = [192, 1024, 1, 1]
+    dtype = "torch.float32"
+    device = "cpu"
+    mean = -0.002
+    std = 0.029
+    data = None
+
+
+class Program_weight_tensor_meta_L_self_modules_features_modules_17_modules_branch2_modules_0_modules_bn_buffers_running_mean_:
+    name = "L_self_modules_features_modules_17_modules_branch2_modules_0_modules_bn_buffers_running_mean_"
+    shape = [192]
+    dtype = "torch.float32"
+    device = "cpu"
+    mean = -0.212
+    std = 0.325
+    data = None
+
+
+class Program_weight_tensor_meta_L_self_modules_features_modules_17_modules_branch2_modules_0_modules_bn_buffers_running_var_:
+    name = "L_self_modules_features_modules_17_modules_branch2_modules_0_modules_bn_buffers_running_var_"
+    shape = [192]
+    dtype = "torch.float32"
+    device = "cpu"
+    mean = 0.232
+    std = 0.110
+    data = None
+
+
+class Program_weight_tensor_meta_L_self_modules_features_modules_17_modules_branch2_modules_0_modules_bn_parameters_weight_:
+    name = "L_self_modules_features_modules_17_modules_branch2_modules_0_modules_bn_parameters_weight_"
+    shape = [192]
+    dtype = "torch.float32"
+    device = "cpu"
+    mean = 1.000
+    std = 0.000
+    data = None
+
+
+class Program_weight_tensor_meta_L_self_modules_features_modules_17_modules_branch2_modules_0_modules_bn_parameters_bias_:
+    name = "L_self_modules_features_modules_17_modules_branch2_modules_0_modules_bn_parameters_bias_"
+    shape = [192]
+    dtype = "torch.float32"
+    device = "cpu"
+    mean = -0.696
+    std = 0.308
+    data = None
+
+
+class Program_weight_tensor_meta_L_self_modules_features_modules_17_modules_branch2_modules_1_modules_conv_parameters_weight_:
+    name = "L_self_modules_features_modules_17_modules_branch2_modules_1_modules_conv_parameters_weight_"
+    shape = [192, 192, 7, 1]
+    dtype = "torch.float32"
+    device = "cpu"
+    mean = -0.002
+    std = 0.024
+    data = None
+
+
+class Program_weight_tensor_meta_L_self_modules_features_modules_17_modules_branch2_modules_1_modules_bn_buffers_running_mean_:
+    name = "L_self_modules_features_modules_17_modules_branch2_modules_1_modules_bn_buffers_running_mean_"
+    shape = [192]
+    dtype = "torch.float32"
+    device = "cpu"
+    mean = -0.297
+    std = 0.207
+    data = None
+
+
+class Program_weight_tensor_meta_L_self_modules_features_modules_17_modules_branch2_modules_1_modules_bn_buffers_running_var_:
+    name = "L_self_modules_features_modules_17_modules_branch2_modules_1_modules_bn_buffers_running_var_"
+    shape = [192]
+    dtype = "torch.float32"
+    device = "cpu"
+    mean = 0.297
+    std = 0.116
+    data = None
+
+
+class Program_weight_tensor_meta_L_self_modules_features_modules_17_modules_branch2_modules_1_modules_bn_parameters_weight_:
+    name = "L_self_modules_features_modules_17_modules_branch2_modules_1_modules_bn_parameters_weight_"
+    shape = [192]
+    dtype = "torch.float32"
+    device = "cpu"
+    mean = 1.000
+    std = 0.000
+    data = None
+
+
+class Program_weight_tensor_meta_L_self_modules_features_modules_17_modules_branch2_modules_1_modules_bn_parameters_bias_:
+    name = "L_self_modules_features_modules_17_modules_branch2_modules_1_modules_bn_parameters_bias_"
+    shape = [192]
+    dtype = "torch.float32"
+    device = "cpu"
+    mean = -0.649
+    std = 0.318
+    data = None
+
+
+class Program_weight_tensor_meta_L_self_modules_features_modules_17_modules_branch2_modules_2_modules_conv_parameters_weight_:
+    name = "L_self_modules_features_modules_17_modules_branch2_modules_2_modules_conv_parameters_weight_"
+    shape = [224, 192, 1, 7]
+    dtype = "torch.float32"
+    device = "cpu"
+    mean = -0.002
+    std = 0.022
+    data = None
+
+
+class Program_weight_tensor_meta_L_self_modules_features_modules_17_modules_branch2_modules_2_modules_bn_buffers_running_mean_:
+    name = "L_self_modules_features_modules_17_modules_branch2_modules_2_modules_bn_buffers_running_mean_"
+    shape = [224]
+    dtype = "torch.float32"
+    device = "cpu"
+    mean = -0.389
+    std = 0.234
+    data = None
+
+
+class Program_weight_tensor_meta_L_self_modules_features_modules_17_modules_branch2_modules_2_modules_bn_buffers_running_var_:
+    name = "L_self_modules_features_modules_17_modules_branch2_modules_2_modules_bn_buffers_running_var_"
+    shape = [224]
+    dtype = "torch.float32"
+    device = "cpu"
+    mean = 0.291
+    std = 0.144
+    data = None
+
+
+class Program_weight_tensor_meta_L_self_modules_features_modules_17_modules_branch2_modules_2_modules_bn_parameters_weight_:
+    name = "L_self_modules_features_modules_17_modules_branch2_modules_2_modules_bn_parameters_weight_"
+    shape = [224]
+    dtype = "torch.float32"
+    device = "cpu"
+    mean = 1.000
+    std = 0.000
+    data = None
+
+
+class Program_weight_tensor_meta_L_self_modules_features_modules_17_modules_branch2_modules_2_modules_bn_parameters_bias_:
+    name = "L_self_modules_features_modules_17_modules_branch2_modules_2_modules_bn_parameters_bias_"
+    shape = [224]
+    dtype = "torch.float32"
+    device = "cpu"
+    mean = -0.738
+    std = 0.302
+    data = None
+
+
+class Program_weight_tensor_meta_L_self_modules_features_modules_17_modules_branch2_modules_3_modules_conv_parameters_weight_:
+    name = "L_self_modules_features_modules_17_modules_branch2_modules_3_modules_conv_parameters_weight_"
+    shape = [224, 224, 7, 1]
+    dtype = "torch.float32"
+    device = "cpu"
+    mean = -0.002
+    std = 0.021
+    data = None
+
+
+class Program_weight_tensor_meta_L_self_modules_features_modules_17_modules_branch2_modules_3_modules_bn_buffers_running_mean_:
+    name = "L_self_modules_features_modules_17_modules_branch2_modules_3_modules_bn_buffers_running_mean_"
+    shape = [224]
+    dtype = "torch.float32"
+    device = "cpu"
+    mean = -0.339
+    std = 0.269
+    data = None
+
+
+class Program_weight_tensor_meta_L_self_modules_features_modules_17_modules_branch2_modules_3_modules_bn_buffers_running_var_:
+    name = "L_self_modules_features_modules_17_modules_branch2_modules_3_modules_bn_buffers_running_var_"
+    shape = [224]
+    dtype = "torch.float32"
+    device = "cpu"
+    mean = 0.295
+    std = 0.134
+    data = None
+
+
+class Program_weight_tensor_meta_L_self_modules_features_modules_17_modules_branch2_modules_3_modules_bn_parameters_weight_:
+    name = "L_self_modules_features_modules_17_modules_branch2_modules_3_modules_bn_parameters_weight_"
+    shape = [224]
+    dtype = "torch.float32"
+    device = "cpu"
+    mean = 1.000
+    std = 0.000
+    data = None
+
+
+class Program_weight_tensor_meta_L_self_modules_features_modules_17_modules_branch2_modules_3_modules_bn_parameters_bias_:
+    name = "L_self_modules_features_modules_17_modules_branch2_modules_3_modules_bn_parameters_bias_"
+    shape = [224]
+    dtype = "torch.float32"
+    device = "cpu"
+    mean = -0.480
+    std = 0.383
+    data = None
+
+
+class Program_weight_tensor_meta_L_self_modules_features_modules_17_modules_branch2_modules_4_modules_conv_parameters_weight_:
+    name = "L_self_modules_features_modules_17_modules_branch2_modules_4_modules_conv_parameters_weight_"
+    shape = [256, 224, 1, 7]
+    dtype = "torch.float32"
+    device = "cpu"
+    mean = -0.003
+    std = 0.020
+    data = None
+
+
+class Program_weight_tensor_meta_L_self_modules_features_modules_17_modules_branch2_modules_4_modules_bn_buffers_running_mean_:
+    name = "L_self_modules_features_modules_17_modules_branch2_modules_4_modules_bn_buffers_running_mean_"
+    shape = [256]
+    dtype = "torch.float32"
+    device = "cpu"
+    mean = -1.060
+    std = 0.310
+    data = None
+
+
+class Program_weight_tensor_meta_L_self_modules_features_modules_17_modules_branch2_modules_4_modules_bn_buffers_running_var_:
+    name = "L_self_modules_features_modules_17_modules_branch2_modules_4_modules_bn_buffers_running_var_"
+    shape = [256]
+    dtype = "torch.float32"
+    device = "cpu"
+    mean = 0.619
+    std = 0.428
+    data = None
+
+
+class Program_weight_tensor_meta_L_self_modules_features_modules_17_modules_branch2_modules_4_modules_bn_parameters_weight_:
+    name = "L_self_modules_features_modules_17_modules_branch2_modules_4_modules_bn_parameters_weight_"
+    shape = [256]
+    dtype = "torch.float32"
+    device = "cpu"
+    mean = 1.000
+    std = 0.000
+    data = None
+
+
+class Program_weight_tensor_meta_L_self_modules_features_modules_17_modules_branch2_modules_4_modules_bn_parameters_bias_:
+    name = "L_self_modules_features_modules_17_modules_branch2_modules_4_modules_bn_parameters_bias_"
+    shape = [256]
+    dtype = "torch.float32"
+    device = "cpu"
+    mean = -1.407
+    std = 0.335
+    data = None
+
+
+class Program_weight_tensor_meta_L_self_modules_features_modules_17_modules_branch3_modules_1_modules_conv_parameters_weight_:
+    name = "L_self_modules_features_modules_17_modules_branch3_modules_1_modules_conv_parameters_weight_"
+    shape = [128, 1024, 1, 1]
+    dtype = "torch.float32"
+    device = "cpu"
+    mean = -0.003
+    std = 0.037
+    data = None
+
+
+class Program_weight_tensor_meta_L_self_modules_features_modules_17_modules_branch3_modules_1_modules_bn_buffers_running_mean_:
+    name = "L_self_modules_features_modules_17_modules_branch3_modules_1_modules_bn_buffers_running_mean_"
+    shape = [128]
+    dtype = "torch.float32"
+    device = "cpu"
+    mean = -0.312
+    std = 0.155
+    data = None
+
+
+class Program_weight_tensor_meta_L_self_modules_features_modules_17_modules_branch3_modules_1_modules_bn_buffers_running_var_:
+    name = "L_self_modules_features_modules_17_modules_branch3_modules_1_modules_bn_buffers_running_var_"
+    shape = [128]
+    dtype = "torch.float32"
+    device = "cpu"
+    mean = 0.172
+    std = 0.299
+    data = None
+
+
+class Program_weight_tensor_meta_L_self_modules_features_modules_17_modules_branch3_modules_1_modules_bn_parameters_weight_:
+    name = "L_self_modules_features_modules_17_modules_branch3_modules_1_modules_bn_parameters_weight_"
+    shape = [128]
+    dtype = "torch.float32"
+    device = "cpu"
+    mean = 1.000
+    std = 0.000
+    data = None
+
+
+class Program_weight_tensor_meta_L_self_modules_features_modules_17_modules_branch3_modules_1_modules_bn_parameters_bias_:
+    name = "L_self_modules_features_modules_17_modules_branch3_modules_1_modules_bn_parameters_bias_"
+    shape = [128]
+    dtype = "torch.float32"
+    device = "cpu"
+    mean = -1.329
+    std = 0.556
+    data = None
+
+
+class Program_weight_tensor_meta_L_self_modules_features_modules_18_modules_branch0_modules_0_modules_conv_parameters_weight_:
+    name = "L_self_modules_features_modules_18_modules_branch0_modules_0_modules_conv_parameters_weight_"
+    shape = [192, 1024, 1, 1]
+    dtype = "torch.float32"
+    device = "cpu"
+    mean = -0.003
+    std = 0.021
+    data = None
+
+
+class Program_weight_tensor_meta_L_self_modules_features_modules_18_modules_branch0_modules_0_modules_bn_buffers_running_mean_:
+    name = "L_self_modules_features_modules_18_modules_branch0_modules_0_modules_bn_buffers_running_mean_"
+    shape = [192]
+    dtype = "torch.float32"
+    device = "cpu"
+    mean = -0.007
+    std = 0.136
+    data = None
+
+
+class Program_weight_tensor_meta_L_self_modules_features_modules_18_modules_branch0_modules_0_modules_bn_buffers_running_var_:
+    name = "L_self_modules_features_modules_18_modules_branch0_modules_0_modules_bn_buffers_running_var_"
+    shape = [192]
+    dtype = "torch.float32"
+    device = "cpu"
+    mean = 0.044
+    std = 0.031
+    data = None
+
+
+class Program_weight_tensor_meta_L_self_modules_features_modules_18_modules_branch0_modules_0_modules_bn_parameters_weight_:
+    name = "L_self_modules_features_modules_18_modules_branch0_modules_0_modules_bn_parameters_weight_"
+    shape = [192]
+    dtype = "torch.float32"
+    device = "cpu"
+    mean = 1.000
+    std = 0.000
+    data = None
+
+
+class Program_weight_tensor_meta_L_self_modules_features_modules_18_modules_branch0_modules_0_modules_bn_parameters_bias_:
+    name = "L_self_modules_features_modules_18_modules_branch0_modules_0_modules_bn_parameters_bias_"
+    shape = [192]
+    dtype = "torch.float32"
+    device = "cpu"
+    mean = -0.665
+    std = 0.264
+    data = None
+
+
+class Program_weight_tensor_meta_L_self_modules_features_modules_18_modules_branch0_modules_1_modules_conv_parameters_weight_:
+    name = "L_self_modules_features_modules_18_modules_branch0_modules_1_modules_conv_parameters_weight_"
+    shape = [192, 192, 3, 3]
+    dtype = "torch.float32"
+    device = "cpu"
+    mean = -0.002
+    std = 0.013
+    data = None
+
+
+class Program_weight_tensor_meta_L_self_modules_features_modules_18_modules_branch0_modules_1_modules_bn_buffers_running_mean_:
+    name = "L_self_modules_features_modules_18_modules_branch0_modules_1_modules_bn_buffers_running_mean_"
+    shape = [192]
+    dtype = "torch.float32"
+    device = "cpu"
+    mean = -0.141
+    std = 0.197
+    data = None
+
+
+class Program_weight_tensor_meta_L_self_modules_features_modules_18_modules_branch0_modules_1_modules_bn_buffers_running_var_:
+    name = "L_self_modules_features_modules_18_modules_branch0_modules_1_modules_bn_buffers_running_var_"
+    shape = [192]
+    dtype = "torch.float32"
+    device = "cpu"
+    mean = 0.057
+    std = 0.047
+    data = None
+
+
+class Program_weight_tensor_meta_L_self_modules_features_modules_18_modules_branch0_modules_1_modules_bn_parameters_weight_:
+    name = "L_self_modules_features_modules_18_modules_branch0_modules_1_modules_bn_parameters_weight_"
+    shape = [192]
+    dtype = "torch.float32"
+    device = "cpu"
+    mean = 1.000
+    std = 0.000
+    data = None
+
+
+class Program_weight_tensor_meta_L_self_modules_features_modules_18_modules_branch0_modules_1_modules_bn_parameters_bias_:
+    name = "L_self_modules_features_modules_18_modules_branch0_modules_1_modules_bn_parameters_bias_"
+    shape = [192]
+    dtype = "torch.float32"
+    device = "cpu"
+    mean = -0.930
+    std = 0.251
+    data = None
+
+
+class Program_weight_tensor_meta_L_self_modules_features_modules_18_modules_branch1_modules_0_modules_conv_parameters_weight_:
+    name = "L_self_modules_features_modules_18_modules_branch1_modules_0_modules_conv_parameters_weight_"
+    shape = [256, 1024, 1, 1]
+    dtype = "torch.float32"
+    device = "cpu"
+    mean = -0.002
+    std = 0.018
+    data = None
+
+
+class Program_weight_tensor_meta_L_self_modules_features_modules_18_modules_branch1_modules_0_modules_bn_buffers_running_mean_:
+    name = "L_self_modules_features_modules_18_modules_branch1_modules_0_modules_bn_buffers_running_mean_"
+    shape = [256]
+    dtype = "torch.float32"
+    device = "cpu"
+    mean = 0.026
+    std = 0.149
+    data = None
+
+
+class Program_weight_tensor_meta_L_self_modules_features_modules_18_modules_branch1_modules_0_modules_bn_buffers_running_var_:
+    name = "L_self_modules_features_modules_18_modules_branch1_modules_0_modules_bn_buffers_running_var_"
+    shape = [256]
+    dtype = "torch.float32"
+    device = "cpu"
+    mean = 0.040
+    std = 0.056
+    data = None
+
+
+class Program_weight_tensor_meta_L_self_modules_features_modules_18_modules_branch1_modules_0_modules_bn_parameters_weight_:
+    name = "L_self_modules_features_modules_18_modules_branch1_modules_0_modules_bn_parameters_weight_"
+    shape = [256]
+    dtype = "torch.float32"
+    device = "cpu"
+    mean = 1.000
+    std = 0.000
+    data = None
+
+
+class Program_weight_tensor_meta_L_self_modules_features_modules_18_modules_branch1_modules_0_modules_bn_parameters_bias_:
+    name = "L_self_modules_features_modules_18_modules_branch1_modules_0_modules_bn_parameters_bias_"
+    shape = [256]
+    dtype = "torch.float32"
+    device = "cpu"
+    mean = -0.544
+    std = 0.327
+    data = None
+
+
+class Program_weight_tensor_meta_L_self_modules_features_modules_18_modules_branch1_modules_1_modules_conv_parameters_weight_:
+    name = "L_self_modules_features_modules_18_modules_branch1_modules_1_modules_conv_parameters_weight_"
+    shape = [256, 256, 1, 7]
+    dtype = "torch.float32"
+    device = "cpu"
+    mean = -0.001
+    std = 0.011
+    data = None
+
+
+class Program_weight_tensor_meta_L_self_modules_features_modules_18_modules_branch1_modules_1_modules_bn_buffers_running_mean_:
+    name = "L_self_modules_features_modules_18_modules_branch1_modules_1_modules_bn_buffers_running_mean_"
+    shape = [256]
+    dtype = "torch.float32"
+    device = "cpu"
+    mean = -0.188
+    std = 0.218
+    data = None
+
+
+class Program_weight_tensor_meta_L_self_modules_features_modules_18_modules_branch1_modules_1_modules_bn_buffers_running_var_:
+    name = "L_self_modules_features_modules_18_modules_branch1_modules_1_modules_bn_buffers_running_var_"
+    shape = [256]
+    dtype = "torch.float32"
+    device = "cpu"
+    mean = 0.108
+    std = 0.071
+    data = None
+
+
+class Program_weight_tensor_meta_L_self_modules_features_modules_18_modules_branch1_modules_1_modules_bn_parameters_weight_:
+    name = "L_self_modules_features_modules_18_modules_branch1_modules_1_modules_bn_parameters_weight_"
+    shape = [256]
+    dtype = "torch.float32"
+    device = "cpu"
+    mean = 1.000
+    std = 0.000
+    data = None
+
+
+class Program_weight_tensor_meta_L_self_modules_features_modules_18_modules_branch1_modules_1_modules_bn_parameters_bias_:
+    name = "L_self_modules_features_modules_18_modules_branch1_modules_1_modules_bn_parameters_bias_"
+    shape = [256]
+    dtype = "torch.float32"
+    device = "cpu"
+    mean = -0.304
+    std = 0.428
+    data = None
+
+
+class Program_weight_tensor_meta_L_self_modules_features_modules_18_modules_branch1_modules_2_modules_conv_parameters_weight_:
+    name = "L_self_modules_features_modules_18_modules_branch1_modules_2_modules_conv_parameters_weight_"
+    shape = [320, 256, 7, 1]
+    dtype = "torch.float32"
+    device = "cpu"
+    mean = -0.001
+    std = 0.011
+    data = None
+
+
+class Program_weight_tensor_meta_L_self_modules_features_modules_18_modules_branch1_modules_2_modules_bn_buffers_running_mean_:
+    name = "L_self_modules_features_modules_18_modules_branch1_modules_2_modules_bn_buffers_running_mean_"
+    shape = [320]
+    dtype = "torch.float32"
+    device = "cpu"
+    mean = -0.315
+    std = 0.396
+    data = None
+
+
+class Program_weight_tensor_meta_L_self_modules_features_modules_18_modules_branch1_modules_2_modules_bn_buffers_running_var_:
+    name = "L_self_modules_features_modules_18_modules_branch1_modules_2_modules_bn_buffers_running_var_"
+    shape = [320]
+    dtype = "torch.float32"
+    device = "cpu"
+    mean = 0.273
+    std = 0.144
+    data = None
+
+
+class Program_weight_tensor_meta_L_self_modules_features_modules_18_modules_branch1_modules_2_modules_bn_parameters_weight_:
+    name = "L_self_modules_features_modules_18_modules_branch1_modules_2_modules_bn_parameters_weight_"
+    shape = [320]
+    dtype = "torch.float32"
+    device = "cpu"
+    mean = 1.000
+    std = 0.000
+    data = None
+
+
+class Program_weight_tensor_meta_L_self_modules_features_modules_18_modules_branch1_modules_2_modules_bn_parameters_bias_:
+    name = "L_self_modules_features_modules_18_modules_branch1_modules_2_modules_bn_parameters_bias_"
+    shape = [320]
+    dtype = "torch.float32"
+    device = "cpu"
+    mean = -0.341
+    std = 0.373
+    data = None
+
+
+class Program_weight_tensor_meta_L_self_modules_features_modules_18_modules_branch1_modules_3_modules_conv_parameters_weight_:
+    name = "L_self_modules_features_modules_18_modules_branch1_modules_3_modules_conv_parameters_weight_"
+    shape = [320, 320, 3, 3]
+    dtype = "torch.float32"
+    device = "cpu"
+    mean = -0.001
+    std = 0.009
+    data = None
+
+
+class Program_weight_tensor_meta_L_self_modules_features_modules_18_modules_branch1_modules_3_modules_bn_buffers_running_mean_:
+    name = "L_self_modules_features_modules_18_modules_branch1_modules_3_modules_bn_buffers_running_mean_"
+    shape = [320]
+    dtype = "torch.float32"
+    device = "cpu"
+    mean = -0.853
+    std = 0.447
+    data = None
+
+
+class Program_weight_tensor_meta_L_self_modules_features_modules_18_modules_branch1_modules_3_modules_bn_buffers_running_var_:
+    name = "L_self_modules_features_modules_18_modules_branch1_modules_3_modules_bn_buffers_running_var_"
+    shape = [320]
+    dtype = "torch.float32"
+    device = "cpu"
+    mean = 0.857
+    std = 0.698
+    data = None
+
+
+class Program_weight_tensor_meta_L_self_modules_features_modules_18_modules_branch1_modules_3_modules_bn_parameters_weight_:
+    name = "L_self_modules_features_modules_18_modules_branch1_modules_3_modules_bn_parameters_weight_"
+    shape = [320]
+    dtype = "torch.float32"
+    device = "cpu"
+    mean = 1.000
+    std = 0.000
+    data = None
+
+
+class Program_weight_tensor_meta_L_self_modules_features_modules_18_modules_branch1_modules_3_modules_bn_parameters_bias_:
+    name = "L_self_modules_features_modules_18_modules_branch1_modules_3_modules_bn_parameters_bias_"
+    shape = [320]
+    dtype = "torch.float32"
+    device = "cpu"
+    mean = -0.807
+    std = 0.326
+    data = None
+
+
+class Program_weight_tensor_meta_L_self_modules_features_modules_19_modules_branch0_modules_conv_parameters_weight_:
+    name = "L_self_modules_features_modules_19_modules_branch0_modules_conv_parameters_weight_"
+    shape = [256, 1536, 1, 1]
+    dtype = "torch.float32"
+    device = "cpu"
+    mean = -0.001
+    std = 0.024
+    data = None
+
+
+class Program_weight_tensor_meta_L_self_modules_features_modules_19_modules_branch0_modules_bn_buffers_running_mean_:
+    name = "L_self_modules_features_modules_19_modules_branch0_modules_bn_buffers_running_mean_"
+    shape = [256]
+    dtype = "torch.float32"
+    device = "cpu"
+    mean = -0.214
+    std = 0.183
+    data = None
+
+
+class Program_weight_tensor_meta_L_self_modules_features_modules_19_modules_branch0_modules_bn_buffers_running_var_:
+    name = "L_self_modules_features_modules_19_modules_branch0_modules_bn_buffers_running_var_"
+    shape = [256]
+    dtype = "torch.float32"
+    device = "cpu"
+    mean = 0.284
+    std = 0.279
+    data = None
+
+
+class Program_weight_tensor_meta_L_self_modules_features_modules_19_modules_branch0_modules_bn_parameters_weight_:
+    name = "L_self_modules_features_modules_19_modules_branch0_modules_bn_parameters_weight_"
+    shape = [256]
+    dtype = "torch.float32"
+    device = "cpu"
+    mean = 1.000
+    std = 0.000
+    data = None
+
+
+class Program_weight_tensor_meta_L_self_modules_features_modules_19_modules_branch0_modules_bn_parameters_bias_:
+    name = (
+        "L_self_modules_features_modules_19_modules_branch0_modules_bn_parameters_bias_"
+    )
+    shape = [256]
+    dtype = "torch.float32"
+    device = "cpu"
+    mean = -0.647
+    std = 0.196
+    data = None
+
+
+class Program_weight_tensor_meta_L_self_modules_features_modules_19_modules_branch1_0_modules_conv_parameters_weight_:
+    name = "L_self_modules_features_modules_19_modules_branch1_0_modules_conv_parameters_weight_"
+    shape = [384, 1536, 1, 1]
+    dtype = "torch.float32"
+    device = "cpu"
+    mean = -0.001
+    std = 0.023
+    data = None
+
+
+class Program_weight_tensor_meta_L_self_modules_features_modules_19_modules_branch1_0_modules_bn_buffers_running_mean_:
+    name = "L_self_modules_features_modules_19_modules_branch1_0_modules_bn_buffers_running_mean_"
+    shape = [384]
+    dtype = "torch.float32"
+    device = "cpu"
+    mean = -0.209
+    std = 0.106
+    data = None
+
+
+class Program_weight_tensor_meta_L_self_modules_features_modules_19_modules_branch1_0_modules_bn_buffers_running_var_:
+    name = "L_self_modules_features_modules_19_modules_branch1_0_modules_bn_buffers_running_var_"
+    shape = [384]
+    dtype = "torch.float32"
+    device = "cpu"
+    mean = 0.241
+    std = 0.057
+    data = None
+
+
+class Program_weight_tensor_meta_L_self_modules_features_modules_19_modules_branch1_0_modules_bn_parameters_weight_:
+    name = "L_self_modules_features_modules_19_modules_branch1_0_modules_bn_parameters_weight_"
+    shape = [384]
+    dtype = "torch.float32"
+    device = "cpu"
+    mean = 1.000
+    std = 0.000
+    data = None
+
+
+class Program_weight_tensor_meta_L_self_modules_features_modules_19_modules_branch1_0_modules_bn_parameters_bias_:
+    name = "L_self_modules_features_modules_19_modules_branch1_0_modules_bn_parameters_bias_"
+    shape = [384]
+    dtype = "torch.float32"
+    device = "cpu"
+    mean = -0.655
+    std = 0.144
+    data = None
+
+
+class Program_weight_tensor_meta_L_self_modules_features_modules_19_modules_branch1_1a_modules_conv_parameters_weight_:
+    name = "L_self_modules_features_modules_19_modules_branch1_1a_modules_conv_parameters_weight_"
+    shape = [256, 384, 1, 3]
+    dtype = "torch.float32"
+    device = "cpu"
+    mean = -0.002
+    std = 0.019
+    data = None
+
+
+class Program_weight_tensor_meta_L_self_modules_features_modules_19_modules_branch1_1a_modules_bn_buffers_running_mean_:
+    name = "L_self_modules_features_modules_19_modules_branch1_1a_modules_bn_buffers_running_mean_"
+    shape = [256]
+    dtype = "torch.float32"
+    device = "cpu"
+    mean = -0.280
+    std = 0.056
+    data = None
+
+
+class Program_weight_tensor_meta_L_self_modules_features_modules_19_modules_branch1_1a_modules_bn_buffers_running_var_:
+    name = "L_self_modules_features_modules_19_modules_branch1_1a_modules_bn_buffers_running_var_"
+    shape = [256]
+    dtype = "torch.float32"
+    device = "cpu"
+    mean = 0.096
+    std = 0.012
+    data = None
+
+
+class Program_weight_tensor_meta_L_self_modules_features_modules_19_modules_branch1_1a_modules_bn_parameters_weight_:
+    name = "L_self_modules_features_modules_19_modules_branch1_1a_modules_bn_parameters_weight_"
+    shape = [256]
+    dtype = "torch.float32"
+    device = "cpu"
+    mean = 1.000
+    std = 0.000
+    data = None
+
+
+class Program_weight_tensor_meta_L_self_modules_features_modules_19_modules_branch1_1a_modules_bn_parameters_bias_:
+    name = "L_self_modules_features_modules_19_modules_branch1_1a_modules_bn_parameters_bias_"
+    shape = [256]
+    dtype = "torch.float32"
+    device = "cpu"
+    mean = -1.138
+    std = 0.107
+    data = None
+
+
+class Program_weight_tensor_meta_L_self_modules_features_modules_19_modules_branch1_1b_modules_conv_parameters_weight_:
+    name = "L_self_modules_features_modules_19_modules_branch1_1b_modules_conv_parameters_weight_"
+    shape = [256, 384, 3, 1]
+    dtype = "torch.float32"
+    device = "cpu"
+    mean = -0.002
+    std = 0.022
+    data = None
+
+
+class Program_weight_tensor_meta_L_self_modules_features_modules_19_modules_branch1_1b_modules_bn_buffers_running_mean_:
+    name = "L_self_modules_features_modules_19_modules_branch1_1b_modules_bn_buffers_running_mean_"
+    shape = [256]
+    dtype = "torch.float32"
+    device = "cpu"
+    mean = -0.285
+    std = 0.045
+    data = None
+
+
+class Program_weight_tensor_meta_L_self_modules_features_modules_19_modules_branch1_1b_modules_bn_buffers_running_var_:
+    name = "L_self_modules_features_modules_19_modules_branch1_1b_modules_bn_buffers_running_var_"
+    shape = [256]
+    dtype = "torch.float32"
+    device = "cpu"
+    mean = 0.101
+    std = 0.009
+    data = None
+
+
+class Program_weight_tensor_meta_L_self_modules_features_modules_19_modules_branch1_1b_modules_bn_parameters_weight_:
+    name = "L_self_modules_features_modules_19_modules_branch1_1b_modules_bn_parameters_weight_"
+    shape = [256]
+    dtype = "torch.float32"
+    device = "cpu"
+    mean = 1.000
+    std = 0.000
+    data = None
+
+
+class Program_weight_tensor_meta_L_self_modules_features_modules_19_modules_branch1_1b_modules_bn_parameters_bias_:
+    name = "L_self_modules_features_modules_19_modules_branch1_1b_modules_bn_parameters_bias_"
+    shape = [256]
+    dtype = "torch.float32"
+    device = "cpu"
+    mean = -1.014
+    std = 0.093
+    data = None
+
+
+class Program_weight_tensor_meta_L_self_modules_features_modules_19_modules_branch2_0_modules_conv_parameters_weight_:
+    name = "L_self_modules_features_modules_19_modules_branch2_0_modules_conv_parameters_weight_"
+    shape = [384, 1536, 1, 1]
+    dtype = "torch.float32"
+    device = "cpu"
+    mean = -0.001
+    std = 0.018
+    data = None
+
+
+class Program_weight_tensor_meta_L_self_modules_features_modules_19_modules_branch2_0_modules_bn_buffers_running_mean_:
+    name = "L_self_modules_features_modules_19_modules_branch2_0_modules_bn_buffers_running_mean_"
+    shape = [384]
+    dtype = "torch.float32"
+    device = "cpu"
+    mean = -0.155
+    std = 0.179
+    data = None
+
+
+class Program_weight_tensor_meta_L_self_modules_features_modules_19_modules_branch2_0_modules_bn_buffers_running_var_:
+    name = "L_self_modules_features_modules_19_modules_branch2_0_modules_bn_buffers_running_var_"
+    shape = [384]
+    dtype = "torch.float32"
+    device = "cpu"
+    mean = 0.176
+    std = 0.124
+    data = None
+
+
+class Program_weight_tensor_meta_L_self_modules_features_modules_19_modules_branch2_0_modules_bn_parameters_weight_:
+    name = "L_self_modules_features_modules_19_modules_branch2_0_modules_bn_parameters_weight_"
+    shape = [384]
+    dtype = "torch.float32"
+    device = "cpu"
+    mean = 1.000
+    std = 0.000
+    data = None
+
+
+class Program_weight_tensor_meta_L_self_modules_features_modules_19_modules_branch2_0_modules_bn_parameters_bias_:
+    name = "L_self_modules_features_modules_19_modules_branch2_0_modules_bn_parameters_bias_"
+    shape = [384]
+    dtype = "torch.float32"
+    device = "cpu"
+    mean = -0.647
+    std = 0.251
+    data = None
+
+
+class Program_weight_tensor_meta_L_self_modules_features_modules_19_modules_branch2_1_modules_conv_parameters_weight_:
+    name = "L_self_modules_features_modules_19_modules_branch2_1_modules_conv_parameters_weight_"
+    shape = [448, 384, 3, 1]
+    dtype = "torch.float32"
+    device = "cpu"
+    mean = -0.002
+    std = 0.017
+    data = None
+
+
+class Program_weight_tensor_meta_L_self_modules_features_modules_19_modules_branch2_1_modules_bn_buffers_running_mean_:
+    name = "L_self_modules_features_modules_19_modules_branch2_1_modules_bn_buffers_running_mean_"
+    shape = [448]
+    dtype = "torch.float32"
+    device = "cpu"
+    mean = -0.202
+    std = 0.128
+    data = None
+
+
+class Program_weight_tensor_meta_L_self_modules_features_modules_19_modules_branch2_1_modules_bn_buffers_running_var_:
+    name = "L_self_modules_features_modules_19_modules_branch2_1_modules_bn_buffers_running_var_"
+    shape = [448]
+    dtype = "torch.float32"
+    device = "cpu"
+    mean = 0.091
+    std = 0.043
+    data = None
+
+
+class Program_weight_tensor_meta_L_self_modules_features_modules_19_modules_branch2_1_modules_bn_parameters_weight_:
+    name = "L_self_modules_features_modules_19_modules_branch2_1_modules_bn_parameters_weight_"
+    shape = [448]
+    dtype = "torch.float32"
+    device = "cpu"
+    mean = 1.000
+    std = 0.000
+    data = None
+
+
+class Program_weight_tensor_meta_L_self_modules_features_modules_19_modules_branch2_1_modules_bn_parameters_bias_:
+    name = "L_self_modules_features_modules_19_modules_branch2_1_modules_bn_parameters_bias_"
+    shape = [448]
+    dtype = "torch.float32"
+    device = "cpu"
+    mean = -0.566
+    std = 0.202
+    data = None
+
+
+class Program_weight_tensor_meta_L_self_modules_features_modules_19_modules_branch2_2_modules_conv_parameters_weight_:
+    name = "L_self_modules_features_modules_19_modules_branch2_2_modules_conv_parameters_weight_"
+    shape = [512, 448, 1, 3]
+    dtype = "torch.float32"
+    device = "cpu"
+    mean = -0.002
+    std = 0.014
+    data = None
+
+
+class Program_weight_tensor_meta_L_self_modules_features_modules_19_modules_branch2_2_modules_bn_buffers_running_mean_:
+    name = "L_self_modules_features_modules_19_modules_branch2_2_modules_bn_buffers_running_mean_"
+    shape = [512]
+    dtype = "torch.float32"
+    device = "cpu"
+    mean = -0.302
+    std = 0.197
+    data = None
+
+
+class Program_weight_tensor_meta_L_self_modules_features_modules_19_modules_branch2_2_modules_bn_buffers_running_var_:
+    name = "L_self_modules_features_modules_19_modules_branch2_2_modules_bn_buffers_running_var_"
+    shape = [512]
+    dtype = "torch.float32"
+    device = "cpu"
+    mean = 0.137
+    std = 0.060
+    data = None
+
+
+class Program_weight_tensor_meta_L_self_modules_features_modules_19_modules_branch2_2_modules_bn_parameters_weight_:
+    name = "L_self_modules_features_modules_19_modules_branch2_2_modules_bn_parameters_weight_"
+    shape = [512]
+    dtype = "torch.float32"
+    device = "cpu"
+    mean = 1.000
+    std = 0.000
+    data = None
+
+
+class Program_weight_tensor_meta_L_self_modules_features_modules_19_modules_branch2_2_modules_bn_parameters_bias_:
+    name = "L_self_modules_features_modules_19_modules_branch2_2_modules_bn_parameters_bias_"
+    shape = [512]
+    dtype = "torch.float32"
+    device = "cpu"
+    mean = -0.579
+    std = 0.235
+    data = None
+
+
+class Program_weight_tensor_meta_L_self_modules_features_modules_19_modules_branch2_3a_modules_conv_parameters_weight_:
+    name = "L_self_modules_features_modules_19_modules_branch2_3a_modules_conv_parameters_weight_"
+    shape = [256, 512, 1, 3]
+    dtype = "torch.float32"
+    device = "cpu"
+    mean = -0.002
+    std = 0.013
+    data = None
+
+
+class Program_weight_tensor_meta_L_self_modules_features_modules_19_modules_branch2_3a_modules_bn_buffers_running_mean_:
+    name = "L_self_modules_features_modules_19_modules_branch2_3a_modules_bn_buffers_running_mean_"
+    shape = [256]
+    dtype = "torch.float32"
+    device = "cpu"
+    mean = -0.490
+    std = 0.232
+    data = None
+
+
+class Program_weight_tensor_meta_L_self_modules_features_modules_19_modules_branch2_3a_modules_bn_buffers_running_var_:
+    name = "L_self_modules_features_modules_19_modules_branch2_3a_modules_bn_buffers_running_var_"
+    shape = [256]
+    dtype = "torch.float32"
+    device = "cpu"
+    mean = 0.272
+    std = 0.413
+    data = None
+
+
+class Program_weight_tensor_meta_L_self_modules_features_modules_19_modules_branch2_3a_modules_bn_parameters_weight_:
+    name = "L_self_modules_features_modules_19_modules_branch2_3a_modules_bn_parameters_weight_"
+    shape = [256]
+    dtype = "torch.float32"
+    device = "cpu"
+    mean = 1.000
+    std = 0.000
+    data = None
+
+
+class Program_weight_tensor_meta_L_self_modules_features_modules_19_modules_branch2_3a_modules_bn_parameters_bias_:
+    name = "L_self_modules_features_modules_19_modules_branch2_3a_modules_bn_parameters_bias_"
+    shape = [256]
+    dtype = "torch.float32"
+    device = "cpu"
+    mean = -1.028
+    std = 0.375
+    data = None
+
+
+class Program_weight_tensor_meta_L_self_modules_features_modules_19_modules_branch2_3b_modules_conv_parameters_weight_:
+    name = "L_self_modules_features_modules_19_modules_branch2_3b_modules_conv_parameters_weight_"
+    shape = [256, 512, 3, 1]
+    dtype = "torch.float32"
+    device = "cpu"
+    mean = -0.002
+    std = 0.016
+    data = None
+
+
+class Program_weight_tensor_meta_L_self_modules_features_modules_19_modules_branch2_3b_modules_bn_buffers_running_mean_:
+    name = "L_self_modules_features_modules_19_modules_branch2_3b_modules_bn_buffers_running_mean_"
+    shape = [256]
+    dtype = "torch.float32"
+    device = "cpu"
+    mean = -0.539
+    std = 0.187
+    data = None
+
+
+class Program_weight_tensor_meta_L_self_modules_features_modules_19_modules_branch2_3b_modules_bn_buffers_running_var_:
+    name = "L_self_modules_features_modules_19_modules_branch2_3b_modules_bn_buffers_running_var_"
+    shape = [256]
+    dtype = "torch.float32"
+    device = "cpu"
+    mean = 0.265
+    std = 0.249
+    data = None
+
+
+class Program_weight_tensor_meta_L_self_modules_features_modules_19_modules_branch2_3b_modules_bn_parameters_weight_:
+    name = "L_self_modules_features_modules_19_modules_branch2_3b_modules_bn_parameters_weight_"
+    shape = [256]
+    dtype = "torch.float32"
+    device = "cpu"
+    mean = 1.000
+    std = 0.000
+    data = None
+
+
+class Program_weight_tensor_meta_L_self_modules_features_modules_19_modules_branch2_3b_modules_bn_parameters_bias_:
+    name = "L_self_modules_features_modules_19_modules_branch2_3b_modules_bn_parameters_bias_"
+    shape = [256]
+    dtype = "torch.float32"
+    device = "cpu"
+    mean = -1.043
+    std = 0.332
+    data = None
+
+
+class Program_weight_tensor_meta_L_self_modules_features_modules_19_modules_branch3_modules_1_modules_conv_parameters_weight_:
+    name = "L_self_modules_features_modules_19_modules_branch3_modules_1_modules_conv_parameters_weight_"
+    shape = [256, 1536, 1, 1]
+    dtype = "torch.float32"
+    device = "cpu"
+    mean = -0.001
+    std = 0.030
+    data = None
+
+
+class Program_weight_tensor_meta_L_self_modules_features_modules_19_modules_branch3_modules_1_modules_bn_buffers_running_mean_:
+    name = "L_self_modules_features_modules_19_modules_branch3_modules_1_modules_bn_buffers_running_mean_"
+    shape = [256]
+    dtype = "torch.float32"
+    device = "cpu"
+    mean = -0.345
+    std = 0.238
+    data = None
+
+
+class Program_weight_tensor_meta_L_self_modules_features_modules_19_modules_branch3_modules_1_modules_bn_buffers_running_var_:
+    name = "L_self_modules_features_modules_19_modules_branch3_modules_1_modules_bn_buffers_running_var_"
+    shape = [256]
+    dtype = "torch.float32"
+    device = "cpu"
+    mean = 0.477
+    std = 0.243
+    data = None
+
+
+class Program_weight_tensor_meta_L_self_modules_features_modules_19_modules_branch3_modules_1_modules_bn_parameters_weight_:
+    name = "L_self_modules_features_modules_19_modules_branch3_modules_1_modules_bn_parameters_weight_"
+    shape = [256]
+    dtype = "torch.float32"
+    device = "cpu"
+    mean = 1.000
+    std = 0.000
+    data = None
+
+
+class Program_weight_tensor_meta_L_self_modules_features_modules_19_modules_branch3_modules_1_modules_bn_parameters_bias_:
+    name = "L_self_modules_features_modules_19_modules_branch3_modules_1_modules_bn_parameters_bias_"
+    shape = [256]
+    dtype = "torch.float32"
+    device = "cpu"
+    mean = -0.236
+    std = 0.141
+    data = None
+
+
+class Program_weight_tensor_meta_L_self_modules_features_modules_20_modules_branch0_modules_conv_parameters_weight_:
+    name = "L_self_modules_features_modules_20_modules_branch0_modules_conv_parameters_weight_"
+    shape = [256, 1536, 1, 1]
+    dtype = "torch.float32"
+    device = "cpu"
+    mean = -0.003
+    std = 0.023
+    data = None
+
+
+class Program_weight_tensor_meta_L_self_modules_features_modules_20_modules_branch0_modules_bn_buffers_running_mean_:
+    name = "L_self_modules_features_modules_20_modules_branch0_modules_bn_buffers_running_mean_"
+    shape = [256]
+    dtype = "torch.float32"
+    device = "cpu"
+    mean = -0.599
+    std = 0.094
+    data = None
+
+
+class Program_weight_tensor_meta_L_self_modules_features_modules_20_modules_branch0_modules_bn_buffers_running_var_:
+    name = "L_self_modules_features_modules_20_modules_branch0_modules_bn_buffers_running_var_"
+    shape = [256]
+    dtype = "torch.float32"
+    device = "cpu"
+    mean = 0.223
+    std = 0.088
+    data = None
+
+
+class Program_weight_tensor_meta_L_self_modules_features_modules_20_modules_branch0_modules_bn_parameters_weight_:
+    name = "L_self_modules_features_modules_20_modules_branch0_modules_bn_parameters_weight_"
+    shape = [256]
+    dtype = "torch.float32"
+    device = "cpu"
+    mean = 1.000
+    std = 0.000
+    data = None
+
+
+class Program_weight_tensor_meta_L_self_modules_features_modules_20_modules_branch0_modules_bn_parameters_bias_:
+    name = (
+        "L_self_modules_features_modules_20_modules_branch0_modules_bn_parameters_bias_"
+    )
+    shape = [256]
+    dtype = "torch.float32"
+    device = "cpu"
+    mean = -0.711
+    std = 0.160
+    data = None
+
+
+class Program_weight_tensor_meta_L_self_modules_features_modules_20_modules_branch1_0_modules_conv_parameters_weight_:
+    name = "L_self_modules_features_modules_20_modules_branch1_0_modules_conv_parameters_weight_"
+    shape = [384, 1536, 1, 1]
+    dtype = "torch.float32"
+    device = "cpu"
+    mean = -0.002
+    std = 0.024
+    data = None
+
+
+class Program_weight_tensor_meta_L_self_modules_features_modules_20_modules_branch1_0_modules_bn_buffers_running_mean_:
+    name = "L_self_modules_features_modules_20_modules_branch1_0_modules_bn_buffers_running_mean_"
+    shape = [384]
+    dtype = "torch.float32"
+    device = "cpu"
+    mean = -0.422
+    std = 0.193
+    data = None
+
+
+class Program_weight_tensor_meta_L_self_modules_features_modules_20_modules_branch1_0_modules_bn_buffers_running_var_:
+    name = "L_self_modules_features_modules_20_modules_branch1_0_modules_bn_buffers_running_var_"
+    shape = [384]
+    dtype = "torch.float32"
+    device = "cpu"
+    mean = 0.149
+    std = 0.162
+    data = None
+
+
+class Program_weight_tensor_meta_L_self_modules_features_modules_20_modules_branch1_0_modules_bn_parameters_weight_:
+    name = "L_self_modules_features_modules_20_modules_branch1_0_modules_bn_parameters_weight_"
+    shape = [384]
+    dtype = "torch.float32"
+    device = "cpu"
+    mean = 1.000
+    std = 0.000
+    data = None
+
+
+class Program_weight_tensor_meta_L_self_modules_features_modules_20_modules_branch1_0_modules_bn_parameters_bias_:
+    name = "L_self_modules_features_modules_20_modules_branch1_0_modules_bn_parameters_bias_"
+    shape = [384]
+    dtype = "torch.float32"
+    device = "cpu"
+    mean = -0.721
+    std = 0.153
+    data = None
+
+
+class Program_weight_tensor_meta_L_self_modules_features_modules_20_modules_branch1_1a_modules_conv_parameters_weight_:
+    name = "L_self_modules_features_modules_20_modules_branch1_1a_modules_conv_parameters_weight_"
+    shape = [256, 384, 1, 3]
+    dtype = "torch.float32"
+    device = "cpu"
+    mean = -0.004
+    std = 0.019
+    data = None
+
+
+class Program_weight_tensor_meta_L_self_modules_features_modules_20_modules_branch1_1a_modules_bn_buffers_running_mean_:
+    name = "L_self_modules_features_modules_20_modules_branch1_1a_modules_bn_buffers_running_mean_"
+    shape = [256]
+    dtype = "torch.float32"
+    device = "cpu"
+    mean = -0.456
+    std = 0.050
+    data = None
+
+
+class Program_weight_tensor_meta_L_self_modules_features_modules_20_modules_branch1_1a_modules_bn_buffers_running_var_:
+    name = "L_self_modules_features_modules_20_modules_branch1_1a_modules_bn_buffers_running_var_"
+    shape = [256]
+    dtype = "torch.float32"
+    device = "cpu"
+    mean = 0.101
+    std = 0.018
+    data = None
+
+
+class Program_weight_tensor_meta_L_self_modules_features_modules_20_modules_branch1_1a_modules_bn_parameters_weight_:
+    name = "L_self_modules_features_modules_20_modules_branch1_1a_modules_bn_parameters_weight_"
+    shape = [256]
+    dtype = "torch.float32"
+    device = "cpu"
+    mean = 1.000
+    std = 0.000
+    data = None
+
+
+class Program_weight_tensor_meta_L_self_modules_features_modules_20_modules_branch1_1a_modules_bn_parameters_bias_:
+    name = "L_self_modules_features_modules_20_modules_branch1_1a_modules_bn_parameters_bias_"
+    shape = [256]
+    dtype = "torch.float32"
+    device = "cpu"
+    mean = -1.147
+    std = 0.137
+    data = None
+
+
+class Program_weight_tensor_meta_L_self_modules_features_modules_20_modules_branch1_1b_modules_conv_parameters_weight_:
+    name = "L_self_modules_features_modules_20_modules_branch1_1b_modules_conv_parameters_weight_"
+    shape = [256, 384, 3, 1]
+    dtype = "torch.float32"
+    device = "cpu"
+    mean = -0.003
+    std = 0.021
+    data = None
+
+
+class Program_weight_tensor_meta_L_self_modules_features_modules_20_modules_branch1_1b_modules_bn_buffers_running_mean_:
+    name = "L_self_modules_features_modules_20_modules_branch1_1b_modules_bn_buffers_running_mean_"
+    shape = [256]
+    dtype = "torch.float32"
+    device = "cpu"
+    mean = -0.435
+    std = 0.108
+    data = None
+
+
+class Program_weight_tensor_meta_L_self_modules_features_modules_20_modules_branch1_1b_modules_bn_buffers_running_var_:
+    name = "L_self_modules_features_modules_20_modules_branch1_1b_modules_bn_buffers_running_var_"
+    shape = [256]
+    dtype = "torch.float32"
+    device = "cpu"
+    mean = 0.102
+    std = 0.061
+    data = None
+
+
+class Program_weight_tensor_meta_L_self_modules_features_modules_20_modules_branch1_1b_modules_bn_parameters_weight_:
+    name = "L_self_modules_features_modules_20_modules_branch1_1b_modules_bn_parameters_weight_"
+    shape = [256]
+    dtype = "torch.float32"
+    device = "cpu"
+    mean = 1.000
+    std = 0.000
+    data = None
+
+
+class Program_weight_tensor_meta_L_self_modules_features_modules_20_modules_branch1_1b_modules_bn_parameters_bias_:
+    name = "L_self_modules_features_modules_20_modules_branch1_1b_modules_bn_parameters_bias_"
+    shape = [256]
+    dtype = "torch.float32"
+    device = "cpu"
+    mean = -1.112
+    std = 0.178
+    data = None
+
+
+class Program_weight_tensor_meta_L_self_modules_features_modules_20_modules_branch2_0_modules_conv_parameters_weight_:
+    name = "L_self_modules_features_modules_20_modules_branch2_0_modules_conv_parameters_weight_"
+    shape = [384, 1536, 1, 1]
+    dtype = "torch.float32"
+    device = "cpu"
+    mean = -0.003
+    std = 0.021
+    data = None
+
+
+class Program_weight_tensor_meta_L_self_modules_features_modules_20_modules_branch2_0_modules_bn_buffers_running_mean_:
+    name = "L_self_modules_features_modules_20_modules_branch2_0_modules_bn_buffers_running_mean_"
+    shape = [384]
+    dtype = "torch.float32"
+    device = "cpu"
+    mean = -0.463
+    std = 0.230
+    data = None
+
+
+class Program_weight_tensor_meta_L_self_modules_features_modules_20_modules_branch2_0_modules_bn_buffers_running_var_:
+    name = "L_self_modules_features_modules_20_modules_branch2_0_modules_bn_buffers_running_var_"
+    shape = [384]
+    dtype = "torch.float32"
+    device = "cpu"
+    mean = 0.231
+    std = 0.395
+    data = None
+
+
+class Program_weight_tensor_meta_L_self_modules_features_modules_20_modules_branch2_0_modules_bn_parameters_weight_:
+    name = "L_self_modules_features_modules_20_modules_branch2_0_modules_bn_parameters_weight_"
+    shape = [384]
+    dtype = "torch.float32"
+    device = "cpu"
+    mean = 1.000
+    std = 0.000
+    data = None
+
+
+class Program_weight_tensor_meta_L_self_modules_features_modules_20_modules_branch2_0_modules_bn_parameters_bias_:
+    name = "L_self_modules_features_modules_20_modules_branch2_0_modules_bn_parameters_bias_"
+    shape = [384]
+    dtype = "torch.float32"
+    device = "cpu"
+    mean = -0.875
+    std = 0.285
+    data = None
+
+
+class Program_weight_tensor_meta_L_self_modules_features_modules_20_modules_branch2_1_modules_conv_parameters_weight_:
+    name = "L_self_modules_features_modules_20_modules_branch2_1_modules_conv_parameters_weight_"
+    shape = [448, 384, 3, 1]
+    dtype = "torch.float32"
+    device = "cpu"
+    mean = -0.002
+    std = 0.016
+    data = None
+
+
+class Program_weight_tensor_meta_L_self_modules_features_modules_20_modules_branch2_1_modules_bn_buffers_running_mean_:
+    name = "L_self_modules_features_modules_20_modules_branch2_1_modules_bn_buffers_running_mean_"
+    shape = [448]
+    dtype = "torch.float32"
+    device = "cpu"
+    mean = -0.099
+    std = 0.064
+    data = None
+
+
+class Program_weight_tensor_meta_L_self_modules_features_modules_20_modules_branch2_1_modules_bn_buffers_running_var_:
+    name = "L_self_modules_features_modules_20_modules_branch2_1_modules_bn_buffers_running_var_"
+    shape = [448]
+    dtype = "torch.float32"
+    device = "cpu"
+    mean = 0.043
+    std = 0.038
+    data = None
+
+
+class Program_weight_tensor_meta_L_self_modules_features_modules_20_modules_branch2_1_modules_bn_parameters_weight_:
+    name = "L_self_modules_features_modules_20_modules_branch2_1_modules_bn_parameters_weight_"
+    shape = [448]
+    dtype = "torch.float32"
+    device = "cpu"
+    mean = 1.000
+    std = 0.000
+    data = None
+
+
+class Program_weight_tensor_meta_L_self_modules_features_modules_20_modules_branch2_1_modules_bn_parameters_bias_:
+    name = "L_self_modules_features_modules_20_modules_branch2_1_modules_bn_parameters_bias_"
+    shape = [448]
+    dtype = "torch.float32"
+    device = "cpu"
+    mean = -0.415
+    std = 0.222
+    data = None
+
+
+class Program_weight_tensor_meta_L_self_modules_features_modules_20_modules_branch2_2_modules_conv_parameters_weight_:
+    name = "L_self_modules_features_modules_20_modules_branch2_2_modules_conv_parameters_weight_"
+    shape = [512, 448, 1, 3]
+    dtype = "torch.float32"
+    device = "cpu"
+    mean = -0.001
+    std = 0.012
+    data = None
+
+
+class Program_weight_tensor_meta_L_self_modules_features_modules_20_modules_branch2_2_modules_bn_buffers_running_mean_:
+    name = "L_self_modules_features_modules_20_modules_branch2_2_modules_bn_buffers_running_mean_"
+    shape = [512]
+    dtype = "torch.float32"
+    device = "cpu"
+    mean = -0.184
+    std = 0.179
+    data = None
+
+
+class Program_weight_tensor_meta_L_self_modules_features_modules_20_modules_branch2_2_modules_bn_buffers_running_var_:
+    name = "L_self_modules_features_modules_20_modules_branch2_2_modules_bn_buffers_running_var_"
+    shape = [512]
+    dtype = "torch.float32"
+    device = "cpu"
+    mean = 0.093
+    std = 0.044
+    data = None
+
+
+class Program_weight_tensor_meta_L_self_modules_features_modules_20_modules_branch2_2_modules_bn_parameters_weight_:
+    name = "L_self_modules_features_modules_20_modules_branch2_2_modules_bn_parameters_weight_"
+    shape = [512]
+    dtype = "torch.float32"
+    device = "cpu"
+    mean = 1.000
+    std = 0.000
+    data = None
+
+
+class Program_weight_tensor_meta_L_self_modules_features_modules_20_modules_branch2_2_modules_bn_parameters_bias_:
+    name = "L_self_modules_features_modules_20_modules_branch2_2_modules_bn_parameters_bias_"
+    shape = [512]
+    dtype = "torch.float32"
+    device = "cpu"
+    mean = -0.260
+    std = 0.255
+    data = None
+
+
+class Program_weight_tensor_meta_L_self_modules_features_modules_20_modules_branch2_3a_modules_conv_parameters_weight_:
+    name = "L_self_modules_features_modules_20_modules_branch2_3a_modules_conv_parameters_weight_"
+    shape = [256, 512, 1, 3]
+    dtype = "torch.float32"
+    device = "cpu"
+    mean = -0.002
+    std = 0.012
+    data = None
+
+
+class Program_weight_tensor_meta_L_self_modules_features_modules_20_modules_branch2_3a_modules_bn_buffers_running_mean_:
+    name = "L_self_modules_features_modules_20_modules_branch2_3a_modules_bn_buffers_running_mean_"
+    shape = [256]
+    dtype = "torch.float32"
+    device = "cpu"
+    mean = -0.761
+    std = 0.319
+    data = None
+
+
+class Program_weight_tensor_meta_L_self_modules_features_modules_20_modules_branch2_3a_modules_bn_buffers_running_var_:
+    name = "L_self_modules_features_modules_20_modules_branch2_3a_modules_bn_buffers_running_var_"
+    shape = [256]
+    dtype = "torch.float32"
+    device = "cpu"
+    mean = 0.859
+    std = 1.758
+    data = None
+
+
+class Program_weight_tensor_meta_L_self_modules_features_modules_20_modules_branch2_3a_modules_bn_parameters_weight_:
+    name = "L_self_modules_features_modules_20_modules_branch2_3a_modules_bn_parameters_weight_"
+    shape = [256]
+    dtype = "torch.float32"
+    device = "cpu"
+    mean = 1.000
+    std = 0.000
+    data = None
+
+
+class Program_weight_tensor_meta_L_self_modules_features_modules_20_modules_branch2_3a_modules_bn_parameters_bias_:
+    name = "L_self_modules_features_modules_20_modules_branch2_3a_modules_bn_parameters_bias_"
+    shape = [256]
+    dtype = "torch.float32"
+    device = "cpu"
+    mean = -0.591
+    std = 0.267
+    data = None
+
+
+class Program_weight_tensor_meta_L_self_modules_features_modules_20_modules_branch2_3b_modules_conv_parameters_weight_:
+    name = "L_self_modules_features_modules_20_modules_branch2_3b_modules_conv_parameters_weight_"
+    shape = [256, 512, 3, 1]
+    dtype = "torch.float32"
+    device = "cpu"
+    mean = -0.002
+    std = 0.013
+    data = None
+
+
+class Program_weight_tensor_meta_L_self_modules_features_modules_20_modules_branch2_3b_modules_bn_buffers_running_mean_:
+    name = "L_self_modules_features_modules_20_modules_branch2_3b_modules_bn_buffers_running_mean_"
+    shape = [256]
+    dtype = "torch.float32"
+    device = "cpu"
+    mean = -0.760
+    std = 0.262
+    data = None
+
+
+class Program_weight_tensor_meta_L_self_modules_features_modules_20_modules_branch2_3b_modules_bn_buffers_running_var_:
+    name = "L_self_modules_features_modules_20_modules_branch2_3b_modules_bn_buffers_running_var_"
+    shape = [256]
+    dtype = "torch.float32"
+    device = "cpu"
+    mean = 0.639
+    std = 0.797
+    data = None
+
+
+class Program_weight_tensor_meta_L_self_modules_features_modules_20_modules_branch2_3b_modules_bn_parameters_weight_:
+    name = "L_self_modules_features_modules_20_modules_branch2_3b_modules_bn_parameters_weight_"
+    shape = [256]
+    dtype = "torch.float32"
+    device = "cpu"
+    mean = 1.000
+    std = 0.000
+    data = None
+
+
+class Program_weight_tensor_meta_L_self_modules_features_modules_20_modules_branch2_3b_modules_bn_parameters_bias_:
+    name = "L_self_modules_features_modules_20_modules_branch2_3b_modules_bn_parameters_bias_"
+    shape = [256]
+    dtype = "torch.float32"
+    device = "cpu"
+    mean = -0.744
+    std = 0.300
+    data = None
+
+
+class Program_weight_tensor_meta_L_self_modules_features_modules_20_modules_branch3_modules_1_modules_conv_parameters_weight_:
+    name = "L_self_modules_features_modules_20_modules_branch3_modules_1_modules_conv_parameters_weight_"
+    shape = [256, 1536, 1, 1]
+    dtype = "torch.float32"
+    device = "cpu"
+    mean = -0.003
+    std = 0.028
+    data = None
+
+
+class Program_weight_tensor_meta_L_self_modules_features_modules_20_modules_branch3_modules_1_modules_bn_buffers_running_mean_:
+    name = "L_self_modules_features_modules_20_modules_branch3_modules_1_modules_bn_buffers_running_mean_"
+    shape = [256]
+    dtype = "torch.float32"
+    device = "cpu"
+    mean = -0.727
+    std = 0.088
+    data = None
+
+
+class Program_weight_tensor_meta_L_self_modules_features_modules_20_modules_branch3_modules_1_modules_bn_buffers_running_var_:
+    name = "L_self_modules_features_modules_20_modules_branch3_modules_1_modules_bn_buffers_running_var_"
+    shape = [256]
+    dtype = "torch.float32"
+    device = "cpu"
+    mean = 0.288
+    std = 0.214
+    data = None
+
+
+class Program_weight_tensor_meta_L_self_modules_features_modules_20_modules_branch3_modules_1_modules_bn_parameters_weight_:
+    name = "L_self_modules_features_modules_20_modules_branch3_modules_1_modules_bn_parameters_weight_"
+    shape = [256]
+    dtype = "torch.float32"
+    device = "cpu"
+    mean = 1.000
+    std = 0.000
+    data = None
+
+
+class Program_weight_tensor_meta_L_self_modules_features_modules_20_modules_branch3_modules_1_modules_bn_parameters_bias_:
+    name = "L_self_modules_features_modules_20_modules_branch3_modules_1_modules_bn_parameters_bias_"
+    shape = [256]
+    dtype = "torch.float32"
+    device = "cpu"
+    mean = -0.490
+    std = 0.116
+    data = None
+
+
+class Program_weight_tensor_meta_L_self_modules_features_modules_21_modules_branch0_modules_conv_parameters_weight_:
+    name = "L_self_modules_features_modules_21_modules_branch0_modules_conv_parameters_weight_"
+    shape = [256, 1536, 1, 1]
+    dtype = "torch.float32"
+    device = "cpu"
+    mean = -0.000
+    std = 0.015
+    data = None
+
+
+class Program_weight_tensor_meta_L_self_modules_features_modules_21_modules_branch0_modules_bn_buffers_running_mean_:
+    name = "L_self_modules_features_modules_21_modules_branch0_modules_bn_buffers_running_mean_"
+    shape = [256]
+    dtype = "torch.float32"
+    device = "cpu"
+    mean = 0.021
+    std = 0.018
+    data = None
+
+
+class Program_weight_tensor_meta_L_self_modules_features_modules_21_modules_branch0_modules_bn_buffers_running_var_:
+    name = "L_self_modules_features_modules_21_modules_branch0_modules_bn_buffers_running_var_"
+    shape = [256]
+    dtype = "torch.float32"
+    device = "cpu"
+    mean = 0.011
+    std = 0.000
+    data = None
+
+
+class Program_weight_tensor_meta_L_self_modules_features_modules_21_modules_branch0_modules_bn_parameters_weight_:
+    name = "L_self_modules_features_modules_21_modules_branch0_modules_bn_parameters_weight_"
+    shape = [256]
+    dtype = "torch.float32"
+    device = "cpu"
+    mean = 1.000
+    std = 0.000
+    data = None
+
+
+class Program_weight_tensor_meta_L_self_modules_features_modules_21_modules_branch0_modules_bn_parameters_bias_:
+    name = (
+        "L_self_modules_features_modules_21_modules_branch0_modules_bn_parameters_bias_"
+    )
+    shape = [256]
+    dtype = "torch.float32"
+    device = "cpu"
+    mean = -0.144
+    std = 0.048
+    data = None
+
+
+class Program_weight_tensor_meta_L_self_modules_features_modules_21_modules_branch1_0_modules_conv_parameters_weight_:
+    name = "L_self_modules_features_modules_21_modules_branch1_0_modules_conv_parameters_weight_"
+    shape = [384, 1536, 1, 1]
+    dtype = "torch.float32"
+    device = "cpu"
+    mean = -0.003
+    std = 0.022
+    data = None
+
+
+class Program_weight_tensor_meta_L_self_modules_features_modules_21_modules_branch1_0_modules_bn_buffers_running_mean_:
+    name = "L_self_modules_features_modules_21_modules_branch1_0_modules_bn_buffers_running_mean_"
+    shape = [384]
+    dtype = "torch.float32"
+    device = "cpu"
+    mean = -0.223
+    std = 0.095
+    data = None
+
+
+class Program_weight_tensor_meta_L_self_modules_features_modules_21_modules_branch1_0_modules_bn_buffers_running_var_:
+    name = "L_self_modules_features_modules_21_modules_branch1_0_modules_bn_buffers_running_var_"
+    shape = [384]
+    dtype = "torch.float32"
+    device = "cpu"
+    mean = 0.052
+    std = 0.068
+    data = None
+
+
+class Program_weight_tensor_meta_L_self_modules_features_modules_21_modules_branch1_0_modules_bn_parameters_weight_:
+    name = "L_self_modules_features_modules_21_modules_branch1_0_modules_bn_parameters_weight_"
+    shape = [384]
+    dtype = "torch.float32"
+    device = "cpu"
+    mean = 1.000
+    std = 0.000
+    data = None
+
+
+class Program_weight_tensor_meta_L_self_modules_features_modules_21_modules_branch1_0_modules_bn_parameters_bias_:
+    name = "L_self_modules_features_modules_21_modules_branch1_0_modules_bn_parameters_bias_"
+    shape = [384]
+    dtype = "torch.float32"
+    device = "cpu"
+    mean = -1.009
+    std = 0.253
+    data = None
+
+
+class Program_weight_tensor_meta_L_self_modules_features_modules_21_modules_branch1_1a_modules_conv_parameters_weight_:
+    name = "L_self_modules_features_modules_21_modules_branch1_1a_modules_conv_parameters_weight_"
+    shape = [256, 384, 1, 3]
+    dtype = "torch.float32"
+    device = "cpu"
+    mean = -0.001
+    std = 0.014
+    data = None
+
+
+class Program_weight_tensor_meta_L_self_modules_features_modules_21_modules_branch1_1a_modules_bn_buffers_running_mean_:
+    name = "L_self_modules_features_modules_21_modules_branch1_1a_modules_bn_buffers_running_mean_"
+    shape = [256]
+    dtype = "torch.float32"
+    device = "cpu"
+    mean = 0.017
+    std = 0.003
+    data = None
+
+
+class Program_weight_tensor_meta_L_self_modules_features_modules_21_modules_branch1_1a_modules_bn_buffers_running_var_:
+    name = "L_self_modules_features_modules_21_modules_branch1_1a_modules_bn_buffers_running_var_"
+    shape = [256]
+    dtype = "torch.float32"
+    device = "cpu"
+    mean = 0.013
+    std = 0.000
+    data = None
+
+
+class Program_weight_tensor_meta_L_self_modules_features_modules_21_modules_branch1_1a_modules_bn_parameters_weight_:
+    name = "L_self_modules_features_modules_21_modules_branch1_1a_modules_bn_parameters_weight_"
+    shape = [256]
+    dtype = "torch.float32"
+    device = "cpu"
+    mean = 1.000
+    std = 0.000
+    data = None
+
+
+class Program_weight_tensor_meta_L_self_modules_features_modules_21_modules_branch1_1a_modules_bn_parameters_bias_:
+    name = "L_self_modules_features_modules_21_modules_branch1_1a_modules_bn_parameters_bias_"
+    shape = [256]
+    dtype = "torch.float32"
+    device = "cpu"
+    mean = -0.092
+    std = 0.035
+    data = None
+
+
+class Program_weight_tensor_meta_L_self_modules_features_modules_21_modules_branch1_1b_modules_conv_parameters_weight_:
+    name = "L_self_modules_features_modules_21_modules_branch1_1b_modules_conv_parameters_weight_"
+    shape = [256, 384, 3, 1]
+    dtype = "torch.float32"
+    device = "cpu"
+    mean = -0.001
+    std = 0.015
+    data = None
+
+
+class Program_weight_tensor_meta_L_self_modules_features_modules_21_modules_branch1_1b_modules_bn_buffers_running_mean_:
+    name = "L_self_modules_features_modules_21_modules_branch1_1b_modules_bn_buffers_running_mean_"
+    shape = [256]
+    dtype = "torch.float32"
+    device = "cpu"
+    mean = 0.018
+    std = 0.003
+    data = None
+
+
+class Program_weight_tensor_meta_L_self_modules_features_modules_21_modules_branch1_1b_modules_bn_buffers_running_var_:
+    name = "L_self_modules_features_modules_21_modules_branch1_1b_modules_bn_buffers_running_var_"
+    shape = [256]
+    dtype = "torch.float32"
+    device = "cpu"
+    mean = 0.013
+    std = 0.000
+    data = None
+
+
+class Program_weight_tensor_meta_L_self_modules_features_modules_21_modules_branch1_1b_modules_bn_parameters_weight_:
+    name = "L_self_modules_features_modules_21_modules_branch1_1b_modules_bn_parameters_weight_"
+    shape = [256]
+    dtype = "torch.float32"
+    device = "cpu"
+    mean = 1.000
+    std = 0.000
+    data = None
+
+
+class Program_weight_tensor_meta_L_self_modules_features_modules_21_modules_branch1_1b_modules_bn_parameters_bias_:
+    name = "L_self_modules_features_modules_21_modules_branch1_1b_modules_bn_parameters_bias_"
+    shape = [256]
+    dtype = "torch.float32"
+    device = "cpu"
+    mean = -0.075
+    std = 0.030
+    data = None
+
+
+class Program_weight_tensor_meta_L_self_modules_features_modules_21_modules_branch2_0_modules_conv_parameters_weight_:
+    name = "L_self_modules_features_modules_21_modules_branch2_0_modules_conv_parameters_weight_"
+    shape = [384, 1536, 1, 1]
+    dtype = "torch.float32"
+    device = "cpu"
+    mean = -0.002
+    std = 0.023
+    data = None
+
+
+class Program_weight_tensor_meta_L_self_modules_features_modules_21_modules_branch2_0_modules_bn_buffers_running_mean_:
+    name = "L_self_modules_features_modules_21_modules_branch2_0_modules_bn_buffers_running_mean_"
+    shape = [384]
+    dtype = "torch.float32"
+    device = "cpu"
+    mean = -0.281
+    std = 0.146
+    data = None
+
+
+class Program_weight_tensor_meta_L_self_modules_features_modules_21_modules_branch2_0_modules_bn_buffers_running_var_:
+    name = "L_self_modules_features_modules_21_modules_branch2_0_modules_bn_buffers_running_var_"
+    shape = [384]
+    dtype = "torch.float32"
+    device = "cpu"
+    mean = 0.078
+    std = 0.139
+    data = None
+
+
+class Program_weight_tensor_meta_L_self_modules_features_modules_21_modules_branch2_0_modules_bn_parameters_weight_:
+    name = "L_self_modules_features_modules_21_modules_branch2_0_modules_bn_parameters_weight_"
+    shape = [384]
+    dtype = "torch.float32"
+    device = "cpu"
+    mean = 1.000
+    std = 0.000
+    data = None
+
+
+class Program_weight_tensor_meta_L_self_modules_features_modules_21_modules_branch2_0_modules_bn_parameters_bias_:
+    name = "L_self_modules_features_modules_21_modules_branch2_0_modules_bn_parameters_bias_"
+    shape = [384]
+    dtype = "torch.float32"
+    device = "cpu"
+    mean = -0.628
+    std = 0.202
+    data = None
+
+
+class Program_weight_tensor_meta_L_self_modules_features_modules_21_modules_branch2_1_modules_conv_parameters_weight_:
+    name = "L_self_modules_features_modules_21_modules_branch2_1_modules_conv_parameters_weight_"
+    shape = [448, 384, 3, 1]
+    dtype = "torch.float32"
+    device = "cpu"
+    mean = -0.002
+    std = 0.018
+    data = None
+
+
+class Program_weight_tensor_meta_L_self_modules_features_modules_21_modules_branch2_1_modules_bn_buffers_running_mean_:
+    name = "L_self_modules_features_modules_21_modules_branch2_1_modules_bn_buffers_running_mean_"
+    shape = [448]
+    dtype = "torch.float32"
+    device = "cpu"
+    mean = -0.140
+    std = 0.077
+    data = None
+
+
+class Program_weight_tensor_meta_L_self_modules_features_modules_21_modules_branch2_1_modules_bn_buffers_running_var_:
+    name = "L_self_modules_features_modules_21_modules_branch2_1_modules_bn_buffers_running_var_"
+    shape = [448]
+    dtype = "torch.float32"
+    device = "cpu"
+    mean = 0.073
+    std = 0.129
+    data = None
+
+
+class Program_weight_tensor_meta_L_self_modules_features_modules_21_modules_branch2_1_modules_bn_parameters_weight_:
+    name = "L_self_modules_features_modules_21_modules_branch2_1_modules_bn_parameters_weight_"
+    shape = [448]
+    dtype = "torch.float32"
+    device = "cpu"
+    mean = 1.000
+    std = 0.000
+    data = None
+
+
+class Program_weight_tensor_meta_L_self_modules_features_modules_21_modules_branch2_1_modules_bn_parameters_bias_:
+    name = "L_self_modules_features_modules_21_modules_branch2_1_modules_bn_parameters_bias_"
+    shape = [448]
+    dtype = "torch.float32"
+    device = "cpu"
+    mean = -0.573
+    std = 0.260
+    data = None
+
+
+class Program_weight_tensor_meta_L_self_modules_features_modules_21_modules_branch2_2_modules_conv_parameters_weight_:
+    name = "L_self_modules_features_modules_21_modules_branch2_2_modules_conv_parameters_weight_"
+    shape = [512, 448, 1, 3]
+    dtype = "torch.float32"
+    device = "cpu"
+    mean = -0.003
+    std = 0.014
+    data = None
+
+
+class Program_weight_tensor_meta_L_self_modules_features_modules_21_modules_branch2_2_modules_bn_buffers_running_mean_:
+    name = "L_self_modules_features_modules_21_modules_branch2_2_modules_bn_buffers_running_mean_"
+    shape = [512]
+    dtype = "torch.float32"
+    device = "cpu"
+    mean = -0.365
+    std = 0.096
+    data = None
+
+
+class Program_weight_tensor_meta_L_self_modules_features_modules_21_modules_branch2_2_modules_bn_buffers_running_var_:
+    name = "L_self_modules_features_modules_21_modules_branch2_2_modules_bn_buffers_running_var_"
+    shape = [512]
+    dtype = "torch.float32"
+    device = "cpu"
+    mean = 0.112
+    std = 0.103
+    data = None
+
+
+class Program_weight_tensor_meta_L_self_modules_features_modules_21_modules_branch2_2_modules_bn_parameters_weight_:
+    name = "L_self_modules_features_modules_21_modules_branch2_2_modules_bn_parameters_weight_"
+    shape = [512]
+    dtype = "torch.float32"
+    device = "cpu"
+    mean = 1.000
+    std = 0.000
+    data = None
+
+
+class Program_weight_tensor_meta_L_self_modules_features_modules_21_modules_branch2_2_modules_bn_parameters_bias_:
+    name = "L_self_modules_features_modules_21_modules_branch2_2_modules_bn_parameters_bias_"
+    shape = [512]
+    dtype = "torch.float32"
+    device = "cpu"
+    mean = -0.708
+    std = 0.200
+    data = None
+
+
+class Program_weight_tensor_meta_L_self_modules_features_modules_21_modules_branch2_3a_modules_conv_parameters_weight_:
+    name = "L_self_modules_features_modules_21_modules_branch2_3a_modules_conv_parameters_weight_"
+    shape = [256, 512, 1, 3]
+    dtype = "torch.float32"
+    device = "cpu"
+    mean = -0.001
+    std = 0.012
+    data = None
+
+
+class Program_weight_tensor_meta_L_self_modules_features_modules_21_modules_branch2_3a_modules_bn_buffers_running_mean_:
+    name = "L_self_modules_features_modules_21_modules_branch2_3a_modules_bn_buffers_running_mean_"
+    shape = [256]
+    dtype = "torch.float32"
+    device = "cpu"
+    mean = -0.057
+    std = 0.019
+    data = None
+
+
+class Program_weight_tensor_meta_L_self_modules_features_modules_21_modules_branch2_3a_modules_bn_buffers_running_var_:
+    name = "L_self_modules_features_modules_21_modules_branch2_3a_modules_bn_buffers_running_var_"
+    shape = [256]
+    dtype = "torch.float32"
+    device = "cpu"
+    mean = 0.041
+    std = 0.002
+    data = None
+
+
+class Program_weight_tensor_meta_L_self_modules_features_modules_21_modules_branch2_3a_modules_bn_parameters_weight_:
+    name = "L_self_modules_features_modules_21_modules_branch2_3a_modules_bn_parameters_weight_"
+    shape = [256]
+    dtype = "torch.float32"
+    device = "cpu"
+    mean = 1.000
+    std = 0.000
+    data = None
+
+
+class Program_weight_tensor_meta_L_self_modules_features_modules_21_modules_branch2_3a_modules_bn_parameters_bias_:
+    name = "L_self_modules_features_modules_21_modules_branch2_3a_modules_bn_parameters_bias_"
+    shape = [256]
+    dtype = "torch.float32"
+    device = "cpu"
+    mean = -0.217
+    std = 0.063
+    data = None
+
+
+class Program_weight_tensor_meta_L_self_modules_features_modules_21_modules_branch2_3b_modules_conv_parameters_weight_:
+    name = "L_self_modules_features_modules_21_modules_branch2_3b_modules_conv_parameters_weight_"
+    shape = [256, 512, 3, 1]
+    dtype = "torch.float32"
+    device = "cpu"
+    mean = -0.001
+    std = 0.012
+    data = None
+
+
+class Program_weight_tensor_meta_L_self_modules_features_modules_21_modules_branch2_3b_modules_bn_buffers_running_mean_:
+    name = "L_self_modules_features_modules_21_modules_branch2_3b_modules_bn_buffers_running_mean_"
+    shape = [256]
+    dtype = "torch.float32"
+    device = "cpu"
+    mean = -0.054
+    std = 0.017
+    data = None
+
+
+class Program_weight_tensor_meta_L_self_modules_features_modules_21_modules_branch2_3b_modules_bn_buffers_running_var_:
+    name = "L_self_modules_features_modules_21_modules_branch2_3b_modules_bn_buffers_running_var_"
+    shape = [256]
+    dtype = "torch.float32"
+    device = "cpu"
+    mean = 0.042
+    std = 0.002
+    data = None
+
+
+class Program_weight_tensor_meta_L_self_modules_features_modules_21_modules_branch2_3b_modules_bn_parameters_weight_:
+    name = "L_self_modules_features_modules_21_modules_branch2_3b_modules_bn_parameters_weight_"
+    shape = [256]
+    dtype = "torch.float32"
+    device = "cpu"
+    mean = 1.000
+    std = 0.000
+    data = None
+
+
+class Program_weight_tensor_meta_L_self_modules_features_modules_21_modules_branch2_3b_modules_bn_parameters_bias_:
+    name = "L_self_modules_features_modules_21_modules_branch2_3b_modules_bn_parameters_bias_"
+    shape = [256]
+    dtype = "torch.float32"
+    device = "cpu"
+    mean = -0.193
+    std = 0.065
+    data = None
+
+
+class Program_weight_tensor_meta_L_self_modules_features_modules_21_modules_branch3_modules_1_modules_conv_parameters_weight_:
+    name = "L_self_modules_features_modules_21_modules_branch3_modules_1_modules_conv_parameters_weight_"
+    shape = [256, 1536, 1, 1]
+    dtype = "torch.float32"
+    device = "cpu"
+    mean = -0.001
+    std = 0.019
+    data = None
+
+
+class Program_weight_tensor_meta_L_self_modules_features_modules_21_modules_branch3_modules_1_modules_bn_buffers_running_mean_:
+    name = "L_self_modules_features_modules_21_modules_branch3_modules_1_modules_bn_buffers_running_mean_"
+    shape = [256]
+    dtype = "torch.float32"
+    device = "cpu"
+    mean = 0.031
+    std = 0.028
+    data = None
+
+
+class Program_weight_tensor_meta_L_self_modules_features_modules_21_modules_branch3_modules_1_modules_bn_buffers_running_var_:
+    name = "L_self_modules_features_modules_21_modules_branch3_modules_1_modules_bn_buffers_running_var_"
+    shape = [256]
+    dtype = "torch.float32"
+    device = "cpu"
+    mean = 0.009
+    std = 0.000
+    data = None
+
+
+class Program_weight_tensor_meta_L_self_modules_features_modules_21_modules_branch3_modules_1_modules_bn_parameters_weight_:
+    name = "L_self_modules_features_modules_21_modules_branch3_modules_1_modules_bn_parameters_weight_"
+    shape = [256]
+    dtype = "torch.float32"
+    device = "cpu"
+    mean = 1.000
+    std = 0.000
+    data = None
+
+
+class Program_weight_tensor_meta_L_self_modules_features_modules_21_modules_branch3_modules_1_modules_bn_parameters_bias_:
+    name = "L_self_modules_features_modules_21_modules_branch3_modules_1_modules_bn_parameters_bias_"
+    shape = [256]
+    dtype = "torch.float32"
+    device = "cpu"
+    mean = -0.043
+    std = 0.041
+    data = None
+
+
+class Program_weight_tensor_meta_L_self_modules_last_linear_parameters_weight_:
+    name = "L_self_modules_last_linear_parameters_weight_"
+    shape = [1000, 1536]
+    dtype = "torch.float32"
+    device = "cpu"
+    mean = -0.000
+    std = 0.035
+    data = None
+
+
+class Program_weight_tensor_meta_L_self_modules_last_linear_parameters_bias_:
+    name = "L_self_modules_last_linear_parameters_bias_"
+    shape = [1000]
+    dtype = "torch.float32"
+    device = "cpu"
+    mean = -0.000
+    std = 0.025
+    data = None