worse test results with your step2 weight

[Haoyan-Guan]

Thanks for your wonderful works. I met a problem that the test results were very bad when using your pre-trained step2 weight without any training. A possible preblem is that the model parameter names are different from the step2 weight you provided. Could you give me some solutions? Part of code output is below,
[04/03 17:52:35 fvcore.common.checkpoint]: [Checkpointer] Loading from /weight/pascal/step2/0/model_best.pth ...
WARNING [04/03 17:52:35 fvcore.common.checkpoint]: Some model parameters or buffers are not found in the checkpoint:
layer0.0.conv_U.weight
layer0.0.conv_V.weight
layer0.0.vector_S
layer0.1.{bias, running_mean, running_var, weight}
layer0.3.conv_U.weight
layer0.3.conv_V.weight
layer0.3.vector_S
layer0.4.{bias, running_mean, running_var, weight}
layer0.6.conv_U.weight
layer0.6.conv_V.weight
layer0.6.vector_S
layer0.7.{bias, running_mean, running_var, weight}
layer1.0.bn1.{bias, running_mean, running_var, weight}
layer1.0.bn2.{bias, running_mean, running_var, weight}
layer1.0.bn3.{bias, running_mean, running_var, weight}
layer1.0.conv1.conv_U.weight
layer1.0.conv1.conv_V.weight
layer1.0.conv1.vector_S
layer1.0.conv2.conv_U.weight
layer1.0.conv2.conv_V.weight
layer1.0.conv2.vector_S
layer1.0.conv3.conv_U.weight
layer1.0.conv3.conv_V.weight
layer1.0.conv3.vector_S
layer1.0.downsample.0.conv_U.weight
layer1.0.downsample.0.conv_V.weight
layer1.0.downsample.0.vector_S
layer1.0.downsample.1.{bias, running_mean, running_var, weight}
layer1.1.bn1.{bias, running_mean, running_var, weight}
layer1.1.bn2.{bias, running_mean, running_var, weight}
layer1.1.bn3.{bias, running_mean, running_var, weight}
layer1.1.conv1.conv_U.weight
layer1.1.conv1.conv_V.weight
layer1.1.conv1.vector_S
layer1.1.conv2.conv_U.weight
layer1.1.conv2.conv_V.weight
layer1.1.conv2.vector_S
layer1.1.conv3.conv_U.weight
layer1.1.conv3.conv_V.weight
layer1.1.conv3.vector_S
layer1.2.bn1.{bias, running_mean, running_var, weight}
layer1.2.bn2.{bias, running_mean, running_var, weight}
layer1.2.bn3.{bias, running_mean, running_var, weight}
layer1.2.conv1.conv_U.weight
layer1.2.conv1.conv_V.weight
layer1.2.conv1.vector_S
layer1.2.conv2.conv_U.weight
layer1.2.conv2.conv_V.weight
layer1.2.conv2.vector_S
layer1.2.conv3.conv_U.weight
layer1.2.conv3.conv_V.weight
layer1.2.conv3.vector_S
layer2.0.bn1.{bias, running_mean, running_var, weight}
layer2.0.bn2.{bias, running_mean, running_var, weight}
layer2.0.bn3.{bias, running_mean, running_var, weight}
layer2.0.conv1.conv_U.weight
layer2.0.conv1.conv_V.weight
layer2.0.conv1.vector_S
layer2.0.conv2.conv_U.weight
layer2.0.conv2.conv_V.weight
layer2.0.conv2.vector_S
layer2.0.conv3.conv_U.weight
layer2.0.conv3.conv_V.weight
layer2.0.conv3.vector_S
layer2.0.downsample.0.conv_U.weight
layer2.0.downsample.0.conv_V.weight
layer2.0.downsample.0.vector_S
layer2.0.downsample.1.{bias, running_mean, running_var, weight}
layer2.1.bn1.{bias, running_mean, running_var, weight}
layer2.1.bn2.{bias, running_mean, running_var, weight}
layer2.1.bn3.{bias, running_mean, running_var, weight}
layer2.1.conv1.conv_U.weight
layer2.1.conv1.conv_V.weight
layer2.1.conv1.vector_S
layer2.1.conv2.conv_U.weight
layer2.1.conv2.conv_V.weight
layer2.1.conv2.vector_S
layer2.1.conv3.conv_U.weight
layer2.1.conv3.conv_V.weight
layer2.1.conv3.vector_S
layer2.2.bn1.{bias, running_mean, running_var, weight}
layer2.2.bn2.{bias, running_mean, running_var, weight}
layer2.2.bn3.{bias, running_mean, running_var, weight}
layer2.2.conv1.conv_U.weight
layer2.2.conv1.conv_V.weight
layer2.2.conv1.vector_S
layer2.2.conv2.conv_U.weight
layer2.2.conv2.conv_V.weight
layer2.2.conv2.vector_S
layer2.2.conv3.conv_U.weight
layer2.2.conv3.conv_V.weight
layer2.2.conv3.vector_S
layer2.3.bn1.{bias, running_mean, running_var, weight}
layer2.3.bn2.{bias, running_mean, running_var, weight}
layer2.3.bn3.{bias, running_mean, running_var, weight}
layer2.3.conv1.conv_U.weight
layer2.3.conv1.conv_V.weight
layer2.3.conv1.vector_S
layer2.3.conv2.conv_U.weight
layer2.3.conv2.conv_V.weight
layer2.3.conv2.vector_S
layer2.3.conv3.conv_U.weight
layer2.3.conv3.conv_V.weight
layer2.3.conv3.vector_S
layer3.0.bn1.{bias, running_mean, running_var, weight}
layer3.0.bn2.{bias, running_mean, running_var, weight}
layer3.0.bn3.{bias, running_mean, running_var, weight}
layer3.0.conv1.conv_U.weight
layer3.0.conv1.conv_V.weight
layer3.0.conv1.vector_S
layer3.0.conv2.conv_U.weight
layer3.0.conv2.conv_V.weight
layer3.0.conv2.vector_S
layer3.0.conv3.conv_U.weight
layer3.0.conv3.conv_V.weight
layer3.0.conv3.vector_S
layer3.0.downsample.0.conv_U.weight
layer3.0.downsample.0.conv_V.weight
layer3.0.downsample.0.vector_S
layer3.0.downsample.1.{bias, running_mean, running_var, weight}
layer3.1.bn1.{bias, running_mean, running_var, weight}
layer3.1.bn2.{bias, running_mean, running_var, weight}
layer3.1.bn3.{bias, running_mean, running_var, weight}
layer3.1.conv1.conv_U.weight
layer3.1.conv1.conv_V.weight
layer3.1.conv1.vector_S
layer3.1.conv2.conv_U.weight
layer3.1.conv2.conv_V.weight
layer3.1.conv2.vector_S
layer3.1.conv3.conv_U.weight
layer3.1.conv3.conv_V.weight
layer3.1.conv3.vector_S
layer3.2.bn1.{bias, running_mean, running_var, weight}
layer3.2.bn2.{bias, running_mean, running_var, weight}
layer3.2.bn3.{bias, running_mean, running_var, weight}
layer3.2.conv1.conv_U.weight
layer3.2.conv1.conv_V.weight
layer3.2.conv1.vector_S
layer3.2.conv2.conv_U.weight
layer3.2.conv2.conv_V.weight
layer3.2.conv2.vector_S
layer3.2.conv3.conv_U.weight
layer3.2.conv3.conv_V.weight
layer3.2.conv3.vector_S
layer3.3.bn1.{bias, running_mean, running_var, weight}
layer3.3.bn2.{bias, running_mean, running_var, weight}
layer3.3.bn3.{bias, running_mean, running_var, weight}
layer3.3.conv1.conv_U.weight
layer3.3.conv1.conv_V.weight
layer3.3.conv1.vector_S
layer3.3.conv2.conv_U.weight
layer3.3.conv2.conv_V.weight
layer3.3.conv2.vector_S
layer3.3.conv3.conv_U.weight
layer3.3.conv3.conv_V.weight
layer3.3.conv3.vector_S
layer3.4.bn1.{bias, running_mean, running_var, weight}
layer3.4.bn2.{bias, running_mean, running_var, weight}
layer3.4.bn3.{bias, running_mean, running_var, weight}
layer3.4.conv1.conv_U.weight
layer3.4.conv1.conv_V.weight
layer3.4.conv1.vector_S
layer3.4.conv2.conv_U.weight
layer3.4.conv2.conv_V.weight
layer3.4.conv2.vector_S
layer3.4.conv3.conv_U.weight
layer3.4.conv3.conv_V.weight
layer3.4.conv3.vector_S
layer3.5.bn1.{bias, running_mean, running_var, weight}
layer3.5.bn2.{bias, running_mean, running_var, weight}
layer3.5.bn3.{bias, running_mean, running_var, weight}
layer3.5.conv1.conv_U.weight
layer3.5.conv1.conv_V.weight
layer3.5.conv1.vector_S
layer3.5.conv2.conv_U.weight
layer3.5.conv2.conv_V.weight
layer3.5.conv2.vector_S
layer3.5.conv3.conv_U.weight
layer3.5.conv3.conv_V.weight
layer3.5.conv3.vector_S
layer4.0.bn1.{bias, running_mean, running_var, weight}
layer4.0.bn2.{bias, running_mean, running_var, weight}
layer4.0.bn3.{bias, running_mean, running_var, weight}
layer4.0.conv1.conv_U.weight
layer4.0.conv1.conv_V.weight
layer4.0.conv1.vector_S
layer4.0.conv2.conv_U.weight
layer4.0.conv2.conv_V.weight
layer4.0.conv2.vector_S
layer4.0.conv3.conv_U.weight
layer4.0.conv3.conv_V.weight
layer4.0.conv3.vector_S
layer4.0.downsample.0.conv_U.weight
layer4.0.downsample.0.conv_V.weight
layer4.0.downsample.0.vector_S
layer4.0.downsample.1.{bias, running_mean, running_var, weight}
layer4.1.bn1.{bias, running_mean, running_var, weight}
layer4.1.bn2.{bias, running_mean, running_var, weight}
layer4.1.bn3.{bias, running_mean, running_var, weight}
layer4.1.conv1.conv_U.weight
layer4.1.conv1.conv_V.weight
layer4.1.conv1.vector_S
layer4.1.conv2.conv_U.weight
layer4.1.conv2.conv_V.weight
layer4.1.conv2.vector_S
layer4.1.conv3.conv_U.weight
layer4.1.conv3.conv_V.weight
layer4.1.conv3.vector_S
layer4.2.bn1.{bias, running_mean, running_var, weight}
layer4.2.bn2.{bias, running_mean, running_var, weight}
layer4.2.bn3.{bias, running_mean, running_var, weight}
layer4.2.conv1.conv_U.weight
layer4.2.conv1.conv_V.weight
layer4.2.conv1.vector_S
layer4.2.conv2.conv_U.weight
layer4.2.conv2.conv_V.weight
layer4.2.conv2.vector_S
layer4.2.conv3.conv_U.weight
layer4.2.conv3.conv_V.weight
layer4.2.conv3.vector_S
WARNING [04/03 17:52:35 fvcore.common.checkpoint]: The checkpoint state_dict contains keys that are not used by the model:
backbone.stem.conv1.weight
backbone.stem.conv1.norm.{bias, running_mean, running_var, weight}
backbone.res2.0.shortcut.weight
backbone.res2.0.shortcut.norm.{bias, running_mean, running_var, weight}
backbone.res2.0.conv1.weight
backbone.res2.0.conv1.norm.{bias, running_mean, running_var, weight}
backbone.res2.0.conv2.weight
backbone.res2.0.conv2.norm.{bias, running_mean, running_var, weight}
backbone.res2.0.conv3.weight
backbone.res2.0.conv3.norm.{bias, running_mean, running_var, weight}
backbone.res2.1.conv1.weight
backbone.res2.1.conv1.norm.{bias, running_mean, running_var, weight}
backbone.res2.1.conv2.weight
backbone.res2.1.conv2.norm.{bias, running_mean, running_var, weight}
backbone.res2.1.conv3.weight
backbone.res2.1.conv3.norm.{bias, running_mean, running_var, weight}
backbone.res2.2.conv1.weight
backbone.res2.2.conv1.norm.{bias, running_mean, running_var, weight}
backbone.res2.2.conv2.weight
backbone.res2.2.conv2.norm.{bias, running_mean, running_var, weight}
backbone.res2.2.conv3.weight
backbone.res2.2.conv3.norm.{bias, running_mean, running_var, weight}
backbone.res3.0.shortcut.weight
backbone.res3.0.shortcut.norm.{bias, running_mean, running_var, weight}
backbone.res3.0.conv1.weight
backbone.res3.0.conv1.norm.{bias, running_mean, running_var, weight}
backbone.res3.0.conv2.weight
backbone.res3.0.conv2.norm.{bias, running_mean, running_var, weight}
backbone.res3.0.conv3.weight
backbone.res3.0.conv3.norm.{bias, running_mean, running_var, weight}
backbone.res3.1.conv1.weight
backbone.res3.1.conv1.norm.{bias, running_mean, running_var, weight}
backbone.res3.1.conv2.weight
backbone.res3.1.conv2.norm.{bias, running_mean, running_var, weight}
backbone.res3.1.conv3.weight
backbone.res3.1.conv3.norm.{bias, running_mean, running_var, weight}
backbone.res3.2.conv1.weight
backbone.res3.2.conv1.norm.{bias, running_mean, running_var, weight}
backbone.res3.2.conv2.weight
backbone.res3.2.conv2.norm.{bias, running_mean, running_var, weight}
backbone.res3.2.conv3.weight
backbone.res3.2.conv3.norm.{bias, running_mean, running_var, weight}
backbone.res3.3.conv1.weight
backbone.res3.3.conv1.norm.{bias, running_mean, running_var, weight}
backbone.res3.3.conv2.weight
backbone.res3.3.conv2.norm.{bias, running_mean, running_var, weight}
backbone.res3.3.conv3.weight
backbone.res3.3.conv3.norm.{bias, running_mean, running_var, weight}
backbone.res4.0.shortcut.weight
backbone.res4.0.shortcut.norm.{bias, running_mean, running_var, weight}
backbone.res4.0.conv1.weight
backbone.res4.0.conv1.norm.{bias, running_mean, running_var, weight}
backbone.res4.0.conv2.weight
backbone.res4.0.conv2.norm.{bias, running_mean, running_var, weight}
backbone.res4.0.conv3.weight
backbone.res4.0.conv3.norm.{bias, running_mean, running_var, weight}
backbone.res4.1.conv1.weight
backbone.res4.1.conv1.norm.{bias, running_mean, running_var, weight}
backbone.res4.1.conv2.weight
backbone.res4.1.conv2.norm.{bias, running_mean, running_var, weight}
backbone.res4.1.conv3.weight
backbone.res4.1.conv3.norm.{bias, running_mean, running_var, weight}
backbone.res4.2.conv1.weight
backbone.res4.2.conv1.norm.{bias, running_mean, running_var, weight}
backbone.res4.2.conv2.weight
backbone.res4.2.conv2.norm.{bias, running_mean, running_var, weight}
backbone.res4.2.conv3.weight
backbone.res4.2.conv3.norm.{bias, running_mean, running_var, weight}
backbone.res4.3.conv1.weight
backbone.res4.3.conv1.norm.{bias, running_mean, running_var, weight}
backbone.res4.3.conv2.weight
backbone.res4.3.conv2.norm.{bias, running_mean, running_var, weight}
backbone.res4.3.conv3.weight
backbone.res4.3.conv3.norm.{bias, running_mean, running_var, weight}
backbone.res4.4.conv1.weight
backbone.res4.4.conv1.norm.{bias, running_mean, running_var, weight}
backbone.res4.4.conv2.weight
backbone.res4.4.conv2.norm.{bias, running_mean, running_var, weight}
backbone.res4.4.conv3.weight
backbone.res4.4.conv3.norm.{bias, running_mean, running_var, weight}
backbone.res4.5.conv1.weight
backbone.res4.5.conv1.norm.{bias, running_mean, running_var, weight}
backbone.res4.5.conv2.weight
backbone.res4.5.conv2.norm.{bias, running_mean, running_var, weight}
backbone.res4.5.conv3.weight
backbone.res4.5.conv3.norm.{bias, running_mean, running_var, weight}
backbone.res5.0.shortcut.weight
backbone.res5.0.shortcut.norm.{bias, running_mean, running_var, weight}
backbone.res5.0.conv1.weight
backbone.res5.0.conv1.norm.{bias, running_mean, running_var, weight}
backbone.res5.0.conv2.weight
backbone.res5.0.conv2.norm.{bias, running_mean, running_var, weight}
backbone.res5.0.conv3.weight
backbone.res5.0.conv3.norm.{bias, running_mean, running_var, weight}
backbone.res5.1.conv1.weight
backbone.res5.1.conv1.norm.{bias, running_mean, running_var, weight}
backbone.res5.1.conv2.weight
backbone.res5.1.conv2.norm.{bias, running_mean, running_var, weight}
backbone.res5.1.conv3.weight
backbone.res5.1.conv3.norm.{bias, running_mean, running_var, weight}
backbone.res5.2.conv1.weight
backbone.res5.2.conv1.norm.{bias, running_mean, running_var, weight}
backbone.res5.2.conv2.weight
backbone.res5.2.conv2.norm.{bias, running_mean, running_var, weight}
backbone.res5.2.conv3.weight
backbone.res5.2.conv3.norm.{bias, running_mean, running_var, weight}
sem_seg_head.pixel_decoder.crossat_0.transformer_cross_attention_layers.0.multihead_attn.{in_proj_bias, in_proj_weight}
sem_seg_head.pixel_decoder.crossat_0.transformer_cross_attention_layers.0.multihead_attn.out_proj.{bias, weight}
sem_seg_head.pixel_decoder.crossat_0.transformer_cross_attention_layers.0.norm.{bias, weight}
sem_seg_head.pixel_decoder.crossat_0.transformer_cross_attention_layers.1.multihead_attn.{in_proj_bias, in_proj_weight}
sem_seg_head.pixel_decoder.crossat_0.transformer_cross_attention_layers.1.multihead_attn.out_proj.{bias, weight}
sem_seg_head.pixel_decoder.crossat_0.transformer_cross_attention_layers.1.norm.{bias, weight}
sem_seg_head.pixel_decoder.crossat_0.transformer_ffn_layers.0.linear1.{bias, weight}
sem_seg_head.pixel_decoder.crossat_0.transformer_ffn_layers.0.linear2.{bias, weight}
sem_seg_head.pixel_decoder.crossat_0.transformer_ffn_layers.0.norm.{bias, weight}
sem_seg_head.pixel_decoder.crossat_0.transformer_ffn_layers.1.linear1.{bias, weight}
sem_seg_head.pixel_decoder.crossat_0.transformer_ffn_layers.1.linear2.{bias, weight}
sem_seg_head.pixel_decoder.crossat_0.transformer_ffn_layers.1.norm.{bias, weight}
sem_seg_head.pixel_decoder.crossat_1.transformer_cross_attention_layers.0.multihead_attn.{in_proj_bias, in_proj_weight}
sem_seg_head.pixel_decoder.crossat_1.transformer_cross_attention_layers.0.multihead_attn.out_proj.{bias, weight}
sem_seg_head.pixel_decoder.crossat_1.transformer_cross_attention_layers.0.norm.{bias, weight}
sem_seg_head.pixel_decoder.crossat_1.transformer_cross_attention_layers.1.multihead_attn.{in_proj_bias, in_proj_weight}
sem_seg_head.pixel_decoder.crossat_1.transformer_cross_attention_layers.1.multihead_attn.out_proj.{bias, weight}
sem_seg_head.pixel_decoder.crossat_1.transformer_cross_attention_layers.1.norm.{bias, weight}
sem_seg_head.pixel_decoder.crossat_1.transformer_ffn_layers.0.linear1.{bias, weight}
sem_seg_head.pixel_decoder.crossat_1.transformer_ffn_layers.0.linear2.{bias, weight}
sem_seg_head.pixel_decoder.crossat_1.transformer_ffn_layers.0.norm.{bias, weight}
sem_seg_head.pixel_decoder.crossat_1.transformer_ffn_layers.1.linear1.{bias, weight}
sem_seg_head.pixel_decoder.crossat_1.transformer_ffn_layers.1.linear2.{bias, weight}
sem_seg_head.pixel_decoder.crossat_1.transformer_ffn_layers.1.norm.{bias, weight}
sem_seg_head.pixel_decoder.crossat_2.transformer_cross_attention_layers.0.multihead_attn.{in_proj_bias, in_proj_weight}
sem_seg_head.pixel_decoder.crossat_2.transformer_cross_attention_layers.0.multihead_attn.out_proj.{bias, weight}
sem_seg_head.pixel_decoder.crossat_2.transformer_cross_attention_layers.0.norm.{bias, weight}
sem_seg_head.pixel_decoder.crossat_2.transformer_cross_attention_layers.1.multihead_attn.{in_proj_bias, in_proj_weight}
sem_seg_head.pixel_decoder.crossat_2.transformer_cross_attention_layers.1.multihead_attn.out_proj.{bias, weight}
sem_seg_head.pixel_decoder.crossat_2.transformer_cross_attention_layers.1.norm.{bias, weight}
sem_seg_head.pixel_decoder.crossat_2.transformer_ffn_layers.0.linear1.{bias, weight}
sem_seg_head.pixel_decoder.crossat_2.transformer_ffn_layers.0.linear2.{bias, weight}
sem_seg_head.pixel_decoder.crossat_2.transformer_ffn_layers.0.norm.{bias, weight}
sem_seg_head.pixel_decoder.crossat_2.transformer_ffn_layers.1.linear1.{bias, weight}
sem_seg_head.pixel_decoder.crossat_2.transformer_ffn_layers.1.linear2.{bias, weight}
sem_seg_head.pixel_decoder.crossat_2.transformer_ffn_layers.1.norm.{bias, weight}
sem_seg_head.pixel_decoder.crossat_4x.transformer_cross_attention_layers.0.multihead_attn.{in_proj_bias, in_proj_weight}
sem_seg_head.pixel_decoder.crossat_4x.transformer_cross_attention_layers.0.multihead_attn.out_proj.{bias, weight}
sem_seg_head.pixel_decoder.crossat_4x.transformer_cross_attention_layers.0.norm.{bias, weight}
sem_seg_head.pixel_decoder.crossat_4x.transformer_cross_attention_layers.1.multihead_attn.{in_proj_bias, in_proj_weight}
sem_seg_head.pixel_decoder.crossat_4x.transformer_cross_attention_layers.1.multihead_attn.out_proj.{bias, weight}
sem_seg_head.pixel_decoder.crossat_4x.transformer_cross_attention_layers.1.norm.{bias, weight}
sem_seg_head.pixel_decoder.crossat_4x.transformer_ffn_layers.0.linear1.{bias, weight}
sem_seg_head.pixel_decoder.crossat_4x.transformer_ffn_layers.0.linear2.{bias, weight}
sem_seg_head.pixel_decoder.crossat_4x.transformer_ffn_layers.0.norm.{bias, weight}
sem_seg_head.pixel_decoder.crossat_4x.transformer_ffn_layers.1.linear1.{bias, weight}
sem_seg_head.pixel_decoder.crossat_4x.transformer_ffn_layers.1.linear2.{bias, weight}
sem_seg_head.pixel_decoder.crossat_4x.transformer_ffn_layers.1.norm.{bias, weight}