jcwang123 / xboundformer

[TMI 2023] XBound-Former: Toward Cross-scale Boundary Modeling in Transformers
41 stars 4 forks source link

Pretrained weights #10

Open houjunlin opened 7 months ago

houjunlin commented 7 months ago

Dear author, Thank you so much for providing the pretrained weights. I tried to load the pretrained weights (i.e. bat_1_1_0_e6_loss_0_aug_1 and bat_valid_1_1_0_e6_loss_0_aug_1 ) to test images and met the following issue. I wonder which model do these weights corresponding to? The xboundformer model or other models like deeplabv3? Thanks!

RuntimeError: Error(s) in loading state_dict for _SimpleSegmentationModel: Missing key(s) in state_dict: "backbone.patch_embed1.proj.weight", "backbone.patch_embed1.proj.bias", "backbone.patch_embed1.norm.weight", "backbone.patch_embed1.norm.bias", "backbone.patch_embed2.proj.weight", "backbone.patch_embed2.proj.bias", "backbone.patch_embed2.norm.weight", "backbone.patch_embed2.norm.bias", "backbone.patch_embed3.proj.weight", "backbone.patch_embed3.proj.bias", "backbone.patch_embed3.norm.weight", "backbone.patch_embed3.norm.bias", "backbone.patch_embed4.proj.weight", "backbone.patch_embed4.proj.bias", "backbone.patch_embed4.norm.weight", "backbone.patch_embed4.norm.bias", "backbone.block1.0.norm1.weight", "backbone.block1.0.norm1.bias", "backbone.block1.0.attn.q.weight", "backbone.block1.0.attn.q.bias", "backbone.block1.0.attn.kv.weight", "backbone.block1.0.attn.kv.bias", "backbone.block1.0.attn.proj.weight", "backbone.block1.0.attn.proj.bias", "backbone.block1.0.attn.sr.weight", "backbone.block1.0.attn.sr.bias", "backbone.block1.0.attn.norm.weight", "backbone.block1.0.attn.norm.bias", "backbone.block1.0.norm2.weight", "backbone.block1.0.norm2.bias", "backbone.block1.0.mlp.fc1.weight", "backbone.block1.0.mlp.fc1.bias", "backbone.block1.0.mlp.dwconv.dwconv.weight", "backbone.block1.0.mlp.dwconv.dwconv.bias", "backbone.block1.0.mlp.fc2.weight", "backbone.block1.0.mlp.fc2.bias", "backbone.block1.1.norm1.weight", "backbone.block1.1.norm1.bias", "backbone.block1.1.attn.q.weight", "backbone.block1.1.attn.q.bias", "backbone.block1.1.attn.kv.weight", "backbone.block1.1.attn.kv.bias", "backbone.block1.1.attn.proj.weight", "backbone.block1.1.attn.proj.bias", "backbone.block1.1.attn.sr.weight", "backbone.block1.1.attn.sr.bias", "backbone.block1.1.attn.norm.weight", "backbone.block1.1.attn.norm.bias", "backbone.block1.1.norm2.weight", "backbone.block1.1.norm2.bias", "backbone.block1.1.mlp.fc1.weight", "backbone.block1.1.mlp.fc1.bias", "backbone.block1.1.mlp.dwconv.dwconv.weight", "backbone.block1.1.mlp.dwconv.dwconv.bias", "backbone.block1.1.mlp.fc2.weight", "backbone.block1.1.mlp.fc2.bias", "backbone.block1.2.norm1.weight", "backbone.block1.2.norm1.bias", "backbone.block1.2.attn.q.weight", "backbone.block1.2.attn.q.bias", "backbone.block1.2.attn.kv.weight", "backbone.block1.2.attn.kv.bias", "backbone.block1.2.attn.proj.weight", "backbone.block1.2.attn.proj.bias", "backbone.block1.2.attn.sr.weight", "backbone.block1.2.attn.sr.bias", "backbone.block1.2.attn.norm.weight", "backbone.block1.2.attn.norm.bias", "backbone.block1.2.norm2.weight", "backbone.block1.2.norm2.bias", "backbone.block1.2.mlp.fc1.weight", "backbone.block1.2.mlp.fc1.bias", "backbone.block1.2.mlp.dwconv.dwconv.weight", "backbone.block1.2.mlp.dwconv.dwconv.bias", "backbone.block1.2.mlp.fc2.weight", "backbone.block1.2.mlp.fc2.bias", "backbone.norm1.weight", "backbone.norm1.bias", "backbone.block2.0.norm1.weight", "backbone.block2.0.norm1.bias", "backbone.block2.0.attn.q.weight", "backbone.block2.0.attn.q.bias", "backbone.block2.0.attn.kv.weight", "backbone.block2.0.attn.kv.bias", "backbone.block2.0.attn.proj.weight", "backbone.block2.0.attn.proj.bias", "backbone.block2.0.attn.sr.weight", "backbone.block2.0.attn.sr.bias", "backbone.block2.0.attn.norm.weight", "backbone.block2.0.attn.norm.bias", "backbone.block2.0.norm2.weight", "backbone.block2.0.norm2.bias", "backbone.block2.0.mlp.fc1.weight", "backbone.block2.0.mlp.fc1.bias", "backbone.block2.0.mlp.dwconv.dwconv.weight", "backbone.block2.0.mlp.dwconv.dwconv.bias", "backbone.block2.0.mlp.fc2.weight", "backbone.block2.0.mlp.fc2.bias", "backbone.block2.1.norm1.weight", "backbone.block2.1.norm1.bias", "backbone.block2.1.attn.q.weight", "backbone.block2.1.attn.q.bias", "backbone.block2.1.attn.kv.weight", "backbone.block2.1.attn.kv.bias", "backbone.block2.1.attn.proj.weight", "backbone.block2.1.attn.proj.bias", "backbone.block2.1.attn.sr.weight", "backbone.block2.1.attn.sr.bias", "backbone.block2.1.attn.norm.weight", "backbone.block2.1.attn.norm.bias", "backbone.block2.1.norm2.weight", "backbone.block2.1.norm2.bias", "backbone.block2.1.mlp.fc1.weight", "backbone.block2.1.mlp.fc1.bias", "backbone.block2.1.mlp.dwconv.dwconv.weight", "backbone.block2.1.mlp.dwconv.dwconv.bias", "backbone.block2.1.mlp.fc2.weight", "backbone.block2.1.mlp.fc2.bias", "backbone.block2.2.norm1.weight", "backbone.block2.2.norm1.bias", "backbone.block2.2.attn.q.weight", "backbone.block2.2.attn.q.bias", "backbone.block2.2.attn.kv.weight", "backbone.block2.2.attn.kv.bias", "backbone.block2.2.attn.proj.weight", "backbone.block2.2.attn.proj.bias", "backbone.block2.2.attn.sr.weight", "backbone.block2.2.attn.sr.bias", "backbone.block2.2.attn.norm.weight", "backbone.block2.2.attn.norm.bias", "backbone.block2.2.norm2.weight", "backbone.block2.2.norm2.bias", "backbone.block2.2.mlp.fc1.weight", "backbone.block2.2.mlp.fc1.bias", "backbone.block2.2.mlp.dwconv.dwconv.weight", "backbone.block2.2.mlp.dwconv.dwconv.bias", "backbone.block2.2.mlp.fc2.weight", "backbone.block2.2.mlp.fc2.bias", "backbone.block2.3.norm1.weight", "backbone.block2.3.norm1.bias", "backbone.block2.3.attn.q.weight", "backbone.block2.3.attn.q.bias", "backbone.block2.3.attn.kv.weight", "backbone.block2.3.attn.kv.bias", "backbone.block2.3.attn.proj.weight", "backbone.block2.3.attn.proj.bias", "backbone.block2.3.attn.sr.weight", "backbone.block2.3.attn.sr.bias", "backbone.block2.3.attn.norm.weight", "backbone.block2.3.attn.norm.bias", "backbone.block2.3.norm2.weight", "backbone.block2.3.norm2.bias", "backbone.block2.3.mlp.fc1.weight", "backbone.block2.3.mlp.fc1.bias", "backbone.block2.3.mlp.dwconv.dwconv.weight", "backbone.block2.3.mlp.dwconv.dwconv.bias", "backbone.block2.3.mlp.fc2.weight", "backbone.block2.3.mlp.fc2.bias", "backbone.norm2.weight", "backbone.norm2.bias", "backbone.block3.0.norm1.weight", "backbone.block3.0.norm1.bias", "backbone.block3.0.attn.q.weight", "backbone.block3.0.attn.q.bias", "backbone.block3.0.attn.kv.weight", "backbone.block3.0.attn.kv.bias", "backbone.block3.0.attn.proj.weight", "backbone.block3.0.attn.proj.bias", "backbone.block3.0.attn.sr.weight", "backbone.block3.0.attn.sr.bias", "backbone.block3.0.attn.norm.weight", "backbone.block3.0.attn.norm.bias", "backbone.block3.0.norm2.weight", "backbone.block3.0.norm2.bias", "backbone.block3.0.mlp.fc1.weight", "backbone.block3.0.mlp.fc1.bias", "backbone.block3.0.mlp.dwconv.dwconv.weight", "backbone.block3.0.mlp.dwconv.dwconv.bias", "backbone.block3.0.mlp.fc2.weight", "backbone.block3.0.mlp.fc2.bias", "backbone.block3.1.norm1.weight", "backbone.block3.1.norm1.bias", "backbone.block3.1.attn.q.weight", "backbone.block3.1.attn.q.bias", "backbone.block3.1.attn.kv.weight", "backbone.block3.1.attn.kv.bias", "backbone.block3.1.attn.proj.weight", "backbone.block3.1.attn.proj.bias", "backbone.block3.1.attn.sr.weight", "backbone.block3.1.attn.sr.bias", "backbone.block3.1.attn.norm.weight", "backbone.block3.1.attn.norm.bias", "backbone.block3.1.norm2.weight", "backbone.block3.1.norm2.bias", "backbone.block3.1.mlp.fc1.weight", "backbone.block3.1.mlp.fc1.bias", "backbone.block3.1.mlp.dwconv.dwconv.weight", "backbone.block3.1.mlp.dwconv.dwconv.bias", "backbone.block3.1.mlp.fc2.weight", "backbone.block3.1.mlp.fc2.bias", "backbone.block3.2.norm1.weight", "backbone.block3.2.norm1.bias", "backbone.block3.2.attn.q.weight", "backbone.block3.2.attn.q.bias", "backbone.block3.2.attn.kv.weight", "backbone.block3.2.attn.kv.bias", "backbone.block3.2.attn.proj.weight", "backbone.block3.2.attn.proj.bias", "backbone.block3.2.attn.sr.weight", "backbone.block3.2.attn.sr.bias", "backbone.block3.2.attn.norm.weight", "backbone.block3.2.attn.norm.bias", "backbone.block3.2.norm2.weight", "backbone.block3.2.norm2.bias", "backbone.block3.2.mlp.fc1.weight", "backbone.block3.2.mlp.fc1.bias", "backbone.block3.2.mlp.dwconv.dwconv.weight", "backbone.block3.2.mlp.dwconv.dwconv.bias", "backbone.block3.2.mlp.fc2.weight", "backbone.block3.2.mlp.fc2.bias", "backbone.block3.3.norm1.weight", "backbone.block3.3.norm1.bias", "backbone.block3.3.attn.q.weight", "backbone.block3.3.attn.q.bias", "backbone.block3.3.attn.kv.weight", "backbone.block3.3.attn.kv.bias", "backbone.block3.3.attn.proj.weight", "backbone.block3.3.attn.proj.bias", "backbone.block3.3.attn.sr.weight", "backbone.block3.3.attn.sr.bias", "backbone.block3.3.attn.norm.weight", "backbone.block3.3.attn.norm.bias", "backbone.block3.3.norm2.weight", "backbone.block3.3.norm2.bias", "backbone.block3.3.mlp.fc1.weight", "backbone.block3.3.mlp.fc1.bias", "backbone.block3.3.mlp.dwconv.dwconv.weight", "backbone.block3.3.mlp.dwconv.dwconv.bias", "backbone.block3.3.mlp.fc2.weight", "backbone.block3.3.mlp.fc2.bias", "backbone.block3.4.norm1.weight", "backbone.block3.4.norm1.bias", "backbone.block3.4.attn.q.weight", "backbone.block3.4.attn.q.bias", "backbone.block3.4.attn.kv.weight", "backbone.block3.4.attn.kv.bias", "backbone.block3.4.attn.proj.weight", "backbone.block3.4.attn.proj.bias", "backbone.block3.4.attn.sr.weight", "backbone.block3.4.attn.sr.bias", "backbone.block3.4.attn.norm.weight", "backbone.block3.4.attn.norm.bias", "backbone.block3.4.norm2.weight", "backbone.block3.4.norm2.bias", "backbone.block3.4.mlp.fc1.weight", "backbone.block3.4.mlp.fc1.bias", "backbone.block3.4.mlp.dwconv.dwconv.weight", "backbone.block3.4.mlp.dwconv.dwconv.bias", "backbone.block3.4.mlp.fc2.weight", "backbone.block3.4.mlp.fc2.bias", "backbone.block3.5.norm1.weight", "backbone.block3.5.norm1.bias", "backbone.block3.5.attn.q.weight", "backbone.block3.5.attn.q.bias", "backbone.block3.5.attn.kv.weight", "backbone.block3.5.attn.kv.bias", "backbone.block3.5.attn.proj.weight", "backbone.block3.5.attn.proj.bias", "backbone.block3.5.attn.sr.weight", "backbone.block3.5.attn.sr.bias", "backbone.block3.5.attn.norm.weight", "backbone.block3.5.attn.norm.bias", "backbone.block3.5.norm2.weight", "backbone.block3.5.norm2.bias", "backbone.block3.5.mlp.fc1.weight", "backbone.block3.5.mlp.fc1.bias", "backbone.block3.5.mlp.dwconv.dwconv.weight", "backbone.block3.5.mlp.dwconv.dwconv.bias", "backbone.block3.5.mlp.fc2.weight", "backbone.block3.5.mlp.fc2.bias", "backbone.norm3.weight", "backbone.norm3.bias", "backbone.block4.0.norm1.weight", "backbone.block4.0.norm1.bias", "backbone.block4.0.attn.q.weight", "backbone.block4.0.attn.q.bias", "backbone.block4.0.attn.kv.weight", "backbone.block4.0.attn.kv.bias", "backbone.block4.0.attn.proj.weight", "backbone.block4.0.attn.proj.bias", "backbone.block4.0.norm2.weight", "backbone.block4.0.norm2.bias", "backbone.block4.0.mlp.fc1.weight", "backbone.block4.0.mlp.fc1.bias", "backbone.block4.0.mlp.dwconv.dwconv.weight", "backbone.block4.0.mlp.dwconv.dwconv.bias", "backbone.block4.0.mlp.fc2.weight", "backbone.block4.0.mlp.fc2.bias", "backbone.block4.1.norm1.weight", "backbone.block4.1.norm1.bias", "backbone.block4.1.attn.q.weight", "backbone.block4.1.attn.q.bias", "backbone.block4.1.attn.kv.weight", "backbone.block4.1.attn.kv.bias", "backbone.block4.1.attn.proj.weight", "backbone.block4.1.attn.proj.bias", "backbone.block4.1.norm2.weight", "backbone.block4.1.norm2.bias", "backbone.block4.1.mlp.fc1.weight", "backbone.block4.1.mlp.fc1.bias", "backbone.block4.1.mlp.dwconv.dwconv.weight", "backbone.block4.1.mlp.dwconv.dwconv.bias", "backbone.block4.1.mlp.fc2.weight", "backbone.block4.1.mlp.fc2.bias", "backbone.block4.2.norm1.weight", "backbone.block4.2.norm1.bias", "backbone.block4.2.attn.q.weight", "backbone.block4.2.attn.q.bias", "backbone.block4.2.attn.kv.weight", "backbone.block4.2.attn.kv.bias", "backbone.block4.2.attn.proj.weight", "backbone.block4.2.attn.proj.bias", "backbone.block4.2.norm2.weight", "backbone.block4.2.norm2.bias", "backbone.block4.2.mlp.fc1.weight", "backbone.block4.2.mlp.fc1.bias", "backbone.block4.2.mlp.dwconv.dwconv.weight", "backbone.block4.2.mlp.dwconv.dwconv.bias", "backbone.block4.2.mlp.fc2.weight", "backbone.block4.2.mlp.fc2.bias", "backbone.norm4.weight", "backbone.norm4.bias", "classifier.classifier.0.weight", "classifier.classifier.1.weight", "classifier.classifier.1.bias", "classifier.classifier.1.running_mean", "classifier.classifier.1.running_var", "classifier.classifier.3.weight", "classifier.classifier.3.bias", "classifier.classifier1.0.weight", "classifier.classifier1.0.bias", "classifier.classifier2.0.weight", "classifier.classifier2.0.bias", "classifier.classifier3.0.weight", "classifier.classifier3.0.bias", "bat_low.convolution_mapping_1.weight", "bat_low.convolution_mapping_1.bias", "bat_low.convolution_mapping_2.weight", "bat_low.convolution_mapping_2.bias", "bat_low.convolution_mapping_3.weight", "bat_low.convolution_mapping_3.bias", "bat_low.im_ex_boud1.query_embed.weight", "bat_low.im_ex_boud1.pos_embed.row_embed.weight", "bat_low.im_ex_boud1.pos_embed.col_embed.weight", "bat_low.im_ex_boud1.transformer.encoder.layers.0.self_attn.in_proj_weight", "bat_low.im_ex_boud1.transformer.encoder.layers.0.self_attn.in_proj_bias", "bat_low.im_ex_boud1.transformer.encoder.layers.0.self_attn.out_proj.weight", "bat_low.im_ex_boud1.transformer.encoder.layers.0.self_attn.out_proj.bias", "bat_low.im_ex_boud1.transformer.encoder.layers.0.linear1.weight", "bat_low.im_ex_boud1.transformer.encoder.layers.0.linear1.bias", "bat_low.im_ex_boud1.transformer.encoder.layers.0.linear2.weight", "bat_low.im_ex_boud1.transformer.encoder.layers.0.linear2.bias", "bat_low.im_ex_boud1.transformer.encoder.layers.0.norm1.weight", "bat_low.im_ex_boud1.transformer.encoder.layers.0.norm1.bias", "bat_low.im_ex_boud1.transformer.encoder.layers.0.norm2.weight", "bat_low.im_ex_boud1.transformer.encoder.layers.0.norm2.bias", "bat_low.im_ex_boud1.transformer.encoder.layers.0.BAG.0.weight", "bat_low.im_ex_boud1.transformer.encoder.layers.0.BAG.1.weight", "bat_low.im_ex_boud1.transformer.encoder.layers.0.BAG.1.bias", "bat_low.im_ex_boud1.transformer.encoder.layers.0.BAG.1.running_mean", "bat_low.im_ex_boud1.transformer.encoder.layers.0.BAG.1.running_var", "bat_low.im_ex_boud1.transformer.encoder.layers.0.BAG.3.weight", "bat_low.im_ex_boud1.transformer.encoder.layers.0.BAG.4.weight", "bat_low.im_ex_boud1.transformer.encoder.layers.0.BAG.4.bias", "bat_low.im_ex_boud1.transformer.encoder.layers.0.BAG.4.running_mean", "bat_low.im_ex_boud1.transformer.encoder.layers.0.BAG.4.running_var", "bat_low.im_ex_boud1.transformer.encoder.layers.0.BAG.6.weight", "bat_low.im_ex_boud1.transformer.encoder.layers.0.BAG.6.bias", "bat_low.im_ex_boud1.transformer.encoder.layers.1.self_attn.in_proj_weight", "bat_low.im_ex_boud1.transformer.encoder.layers.1.self_attn.in_proj_bias", "bat_low.im_ex_boud1.transformer.encoder.layers.1.self_attn.out_proj.weight", "bat_low.im_ex_boud1.transformer.encoder.layers.1.self_attn.out_proj.bias", "bat_low.im_ex_boud1.transformer.encoder.layers.1.linear1.weight", "bat_low.im_ex_boud1.transformer.encoder.layers.1.linear1.bias", "bat_low.im_ex_boud1.transformer.encoder.layers.1.linear2.weight", "bat_low.im_ex_boud1.transformer.encoder.layers.1.linear2.bias", "bat_low.im_ex_boud1.transformer.encoder.layers.1.norm1.weight", "bat_low.im_ex_boud1.transformer.encoder.layers.1.norm1.bias", "bat_low.im_ex_boud1.transformer.encoder.layers.1.norm2.weight", "bat_low.im_ex_boud1.transformer.encoder.layers.1.norm2.bias", "bat_low.im_ex_boud1.transformer.encoder.layers.1.BAG.0.weight", "bat_low.im_ex_boud1.transformer.encoder.layers.1.BAG.1.weight", "bat_low.im_ex_boud1.transformer.encoder.layers.1.BAG.1.bias", "bat_low.im_ex_boud1.transformer.encoder.layers.1.BAG.1.running_mean", "bat_low.im_ex_boud1.transformer.encoder.layers.1.BAG.1.running_var", "bat_low.im_ex_boud1.transformer.encoder.layers.1.BAG.3.weight", "bat_low.im_ex_boud1.transformer.encoder.layers.1.BAG.4.weight", "bat_low.im_ex_boud1.transformer.encoder.layers.1.BAG.4.bias", "bat_low.im_ex_boud1.transformer.encoder.layers.1.BAG.4.running_mean", "bat_low.im_ex_boud1.transformer.encoder.layers.1.BAG.4.running_var", "bat_low.im_ex_boud1.transformer.encoder.layers.1.BAG.6.weight", "bat_low.im_ex_boud1.transformer.encoder.layers.1.BAG.6.bias", "bat_low.im_ex_boud1.transformer.encoder.norm.weight", "bat_low.im_ex_boud1.transformer.encoder.norm.bias", "bat_low.im_ex_boud1.transformer.decoder.layers.0.self_attn.in_proj_weight", "bat_low.im_ex_boud1.transformer.decoder.layers.0.self_attn.in_proj_bias", "bat_low.im_ex_boud1.transformer.decoder.layers.0.self_attn.out_proj.weight", "bat_low.im_ex_boud1.transformer.decoder.layers.0.self_attn.out_proj.bias", "bat_low.im_ex_boud1.transformer.decoder.layers.0.multihead_attn.in_proj_weight", "bat_low.im_ex_boud1.transformer.decoder.layers.0.multihead_attn.in_proj_bias", "bat_low.im_ex_boud1.transformer.decoder.layers.0.multihead_attn.out_proj.weight", "bat_low.im_ex_boud1.transformer.decoder.layers.0.multihead_attn.out_proj.bias", "bat_low.im_ex_boud1.transformer.decoder.layers.0.linear1.weight", "bat_low.im_ex_boud1.transformer.decoder.layers.0.linear1.bias", "bat_low.im_ex_boud1.transformer.decoder.layers.0.linear2.weight", "bat_low.im_ex_boud1.transformer.decoder.layers.0.linear2.bias", "bat_low.im_ex_boud1.transformer.decoder.layers.0.norm1.weight", "bat_low.im_ex_boud1.transformer.decoder.layers.0.norm1.bias", "bat_low.im_ex_boud1.transformer.decoder.layers.0.norm2.weight", "bat_low.im_ex_boud1.transformer.decoder.layers.0.norm2.bias", "bat_low.im_ex_boud1.transformer.decoder.layers.0.norm3.weight", "bat_low.im_ex_boud1.transformer.decoder.layers.0.norm3.bias", "bat_low.im_ex_boud1.transformer.decoder.layers.1.self_attn.in_proj_weight", "bat_low.im_ex_boud1.transformer.decoder.layers.1.self_attn.in_proj_bias", "bat_low.im_ex_boud1.transformer.decoder.layers.1.self_attn.out_proj.weight", "bat_low.im_ex_boud1.transformer.decoder.layers.1.self_attn.out_proj.bias", "bat_low.im_ex_boud1.transformer.decoder.layers.1.multihead_attn.in_proj_weight", "bat_low.im_ex_boud1.transformer.decoder.layers.1.multihead_attn.in_proj_bias", "bat_low.im_ex_boud1.transformer.decoder.layers.1.multihead_attn.out_proj.weight", "bat_low.im_ex_boud1.transformer.decoder.layers.1.multihead_attn.out_proj.bias", "bat_low.im_ex_boud1.transformer.decoder.layers.1.linear1.weight", "bat_low.im_ex_boud1.transformer.decoder.layers.1.linear1.bias", "bat_low.im_ex_boud1.transformer.decoder.layers.1.linear2.weight", "bat_low.im_ex_boud1.transformer.decoder.layers.1.linear2.bias", "bat_low.im_ex_boud1.transformer.decoder.layers.1.norm1.weight", "bat_low.im_ex_boud1.transformer.decoder.layers.1.norm1.bias", "bat_low.im_ex_boud1.transformer.decoder.layers.1.norm2.weight", "bat_low.im_ex_boud1.transformer.decoder.layers.1.norm2.bias", "bat_low.im_ex_boud1.transformer.decoder.layers.1.norm3.weight", "bat_low.im_ex_boud1.transformer.decoder.layers.1.norm3.bias", "bat_low.im_ex_boud1.transformer.decoder.norm.weight", "bat_low.im_ex_boud1.transformer.decoder.norm.bias", "bat_low.im_ex_boud2.query_embed.weight", "bat_low.im_ex_boud2.pos_embed.row_embed.weight", "bat_low.im_ex_boud2.pos_embed.col_embed.weight", "bat_low.im_ex_boud2.transformer.encoder.layers.0.self_attn.in_proj_weight", "bat_low.im_ex_boud2.transformer.encoder.layers.0.self_attn.in_proj_bias", "bat_low.im_ex_boud2.transformer.encoder.layers.0.self_attn.out_proj.weight", "bat_low.im_ex_boud2.transformer.encoder.layers.0.self_attn.out_proj.bias", "bat_low.im_ex_boud2.transformer.encoder.layers.0.linear1.weight", "bat_low.im_ex_boud2.transformer.encoder.layers.0.linear1.bias", "bat_low.im_ex_boud2.transformer.encoder.layers.0.linear2.weight", "bat_low.im_ex_boud2.transformer.encoder.layers.0.linear2.bias", "bat_low.im_ex_boud2.transformer.encoder.layers.0.norm1.weight", "bat_low.im_ex_boud2.transformer.encoder.layers.0.norm1.bias", "bat_low.im_ex_boud2.transformer.encoder.layers.0.norm2.weight", "bat_low.im_ex_boud2.transformer.encoder.layers.0.norm2.bias", "bat_low.im_ex_boud2.transformer.encoder.layers.0.BAG.0.weight", "bat_low.im_ex_boud2.transformer.encoder.layers.0.BAG.1.weight", "bat_low.im_ex_boud2.transformer.encoder.layers.0.BAG.1.bias", "bat_low.im_ex_boud2.transformer.encoder.layers.0.BAG.1.running_mean", "bat_low.im_ex_boud2.transformer.encoder.layers.0.BAG.1.running_var", "bat_low.im_ex_boud2.transformer.encoder.layers.0.BAG.3.weight", "bat_low.im_ex_boud2.transformer.encoder.layers.0.BAG.4.weight", "bat_low.im_ex_boud2.transformer.encoder.layers.0.BAG.4.bias", "bat_low.im_ex_boud2.transformer.encoder.layers.0.BAG.4.running_mean", "bat_low.im_ex_boud2.transformer.encoder.layers.0.BAG.4.running_var", "bat_low.im_ex_boud2.transformer.encoder.layers.0.BAG.6.weight", "bat_low.im_ex_boud2.transformer.encoder.layers.0.BAG.6.bias", "bat_low.im_ex_boud2.transformer.encoder.layers.1.self_attn.in_proj_weight", "bat_low.im_ex_boud2.transformer.encoder.layers.1.self_attn.in_proj_bias", "bat_low.im_ex_boud2.transformer.encoder.layers.1.self_attn.out_proj.weight", "bat_low.im_ex_boud2.transformer.encoder.layers.1.self_attn.out_proj.bias", "bat_low.im_ex_boud2.transformer.encoder.layers.1.linear1.weight", "bat_low.im_ex_boud2.transformer.encoder.layers.1.linear1.bias", "bat_low.im_ex_boud2.transformer.encoder.layers.1.linear2.weight", "bat_low.im_ex_boud2.transformer.encoder.layers.1.linear2.bias", "bat_low.im_ex_boud2.transformer.encoder.layers.1.norm1.weight", "bat_low.im_ex_boud2.transformer.encoder.layers.1.norm1.bias", "bat_low.im_ex_boud2.transformer.encoder.layers.1.norm2.weight", "bat_low.im_ex_boud2.transformer.encoder.layers.1.norm2.bias", "bat_low.im_ex_boud2.transformer.encoder.layers.1.BAG.0.weight", "bat_low.im_ex_boud2.transformer.encoder.layers.1.BAG.1.weight", "bat_low.im_ex_boud2.transformer.encoder.layers.1.BAG.1.bias", "bat_low.im_ex_boud2.transformer.encoder.layers.1.BAG.1.running_mean", "bat_low.im_ex_boud2.transformer.encoder.layers.1.BAG.1.running_var", "bat_low.im_ex_boud2.transformer.encoder.layers.1.BAG.3.weight", "bat_low.im_ex_boud2.transformer.encoder.layers.1.BAG.4.weight", "bat_low.im_ex_boud2.transformer.encoder.layers.1.BAG.4.bias", "bat_low.im_ex_boud2.transformer.encoder.layers.1.BAG.4.running_mean", "bat_low.im_ex_boud2.transformer.encoder.layers.1.BAG.4.running_var", "bat_low.im_ex_boud2.transformer.encoder.layers.1.BAG.6.weight", "bat_low.im_ex_boud2.transformer.encoder.layers.1.BAG.6.bias", "bat_low.im_ex_boud2.transformer.encoder.norm.weight", "bat_low.im_ex_boud2.transformer.encoder.norm.bias", "bat_low.im_ex_boud2.transformer.decoder.layers.0.self_attn.in_proj_weight", "bat_low.im_ex_boud2.transformer.decoder.layers.0.self_attn.in_proj_bias", "bat_low.im_ex_boud2.transformer.decoder.layers.0.self_attn.out_proj.weight", "bat_low.im_ex_boud2.transformer.decoder.layers.0.self_attn.out_proj.bias", "bat_low.im_ex_boud2.transformer.decoder.layers.0.multihead_attn.in_proj_weight", "bat_low.im_ex_boud2.transformer.decoder.layers.0.multihead_attn.in_proj_bias", "bat_low.im_ex_boud2.transformer.decoder.layers.0.multihead_attn.out_proj.weight", "bat_low.im_ex_boud2.transformer.decoder.layers.0.multihead_attn.out_proj.bias", "bat_low.im_ex_boud2.transformer.decoder.layers.0.linear1.weight", "bat_low.im_ex_boud2.transformer.decoder.layers.0.linear1.bias", "bat_low.im_ex_boud2.transformer.decoder.layers.0.linear2.weight", "bat_low.im_ex_boud2.transformer.decoder.layers.0.linear2.bias", "bat_low.im_ex_boud2.transformer.decoder.layers.0.norm1.weight", "bat_low.im_ex_boud2.transformer.decoder.layers.0.norm1.bias", "bat_low.im_ex_boud2.transformer.decoder.layers.0.norm2.weight", "bat_low.im_ex_boud2.transformer.decoder.layers.0.norm2.bias", "bat_low.im_ex_boud2.transformer.decoder.layers.0.norm3.weight", "bat_low.im_ex_boud2.transformer.decoder.layers.0.norm3.bias", "bat_low.im_ex_boud2.transformer.decoder.layers.1.self_attn.in_proj_weight", "bat_low.im_ex_boud2.transformer.decoder.layers.1.self_attn.in_proj_bias", "bat_low.im_ex_boud2.transformer.decoder.layers.1.self_attn.out_proj.weight", "bat_low.im_ex_boud2.transformer.decoder.layers.1.self_attn.out_proj.bias", "bat_low.im_ex_boud2.transformer.decoder.layers.1.multihead_attn.in_proj_weight", "bat_low.im_ex_boud2.transformer.decoder.layers.1.multihead_attn.in_proj_bias", "bat_low.im_ex_boud2.transformer.decoder.layers.1.multihead_attn.out_proj.weight", "bat_low.im_ex_boud2.transformer.decoder.layers.1.multihead_attn.out_proj.bias", "bat_low.im_ex_boud2.transformer.decoder.layers.1.linear1.weight", "bat_low.im_ex_boud2.transformer.decoder.layers.1.linear1.bias", "bat_low.im_ex_boud2.transformer.decoder.layers.1.linear2.weight", "bat_low.im_ex_boud2.transformer.decoder.layers.1.linear2.bias", "bat_low.im_ex_boud2.transformer.decoder.layers.1.norm1.weight", "bat_low.im_ex_boud2.transformer.decoder.layers.1.norm1.bias", "bat_low.im_ex_boud2.transformer.decoder.layers.1.norm2.weight", "bat_low.im_ex_boud2.transformer.decoder.layers.1.norm2.bias", "bat_low.im_ex_boud2.transformer.decoder.layers.1.norm3.weight", "bat_low.im_ex_boud2.transformer.decoder.layers.1.norm3.bias", "bat_low.im_ex_boud2.transformer.decoder.norm.weight", "bat_low.im_ex_boud2.transformer.decoder.norm.bias", "bat_low.im_ex_boud3.query_embed.weight", "bat_low.im_ex_boud3.pos_embed.row_embed.weight", "bat_low.im_ex_boud3.pos_embed.col_embed.weight", "bat_low.im_ex_boud3.transformer.encoder.layers.0.self_attn.in_proj_weight", "bat_low.im_ex_boud3.transformer.encoder.layers.0.self_attn.in_proj_bias", "bat_low.im_ex_boud3.transformer.encoder.layers.0.self_attn.out_proj.weight", "bat_low.im_ex_boud3.transformer.encoder.layers.0.self_attn.out_proj.bias", "bat_low.im_ex_boud3.transformer.encoder.layers.0.linear1.weight", "bat_low.im_ex_boud3.transformer.encoder.layers.0.linear1.bias", "bat_low.im_ex_boud3.transformer.encoder.layers.0.linear2.weight", "bat_low.im_ex_boud3.transformer.encoder.layers.0.linear2.bias", "bat_low.im_ex_boud3.transformer.encoder.layers.0.norm1.weight", "bat_low.im_ex_boud3.transformer.encoder.layers.0.norm1.bias", "bat_low.im_ex_boud3.transformer.encoder.layers.0.norm2.weight", "bat_low.im_ex_boud3.transformer.encoder.layers.0.norm2.bias", "bat_low.im_ex_boud3.transformer.encoder.layers.0.BAG.0.weight", "bat_low.im_ex_boud3.transformer.encoder.layers.0.BAG.1.weight", "bat_low.im_ex_boud3.transformer.encoder.layers.0.BAG.1.bias", "bat_low.im_ex_boud3.transformer.encoder.layers.0.BAG.1.running_mean", "bat_low.im_ex_boud3.transformer.encoder.layers.0.BAG.1.running_var", "bat_low.im_ex_boud3.transformer.encoder.layers.0.BAG.3.weight", "bat_low.im_ex_boud3.transformer.encoder.layers.0.BAG.4.weight", "bat_low.im_ex_boud3.transformer.encoder.layers.0.BAG.4.bias", "bat_low.im_ex_boud3.transformer.encoder.layers.0.BAG.4.running_mean", "bat_low.im_ex_boud3.transformer.encoder.layers.0.BAG.4.running_var", "bat_low.im_ex_boud3.transformer.encoder.layers.0.BAG.6.weight", "bat_low.im_ex_boud3.transformer.encoder.layers.0.BAG.6.bias", "bat_low.im_ex_boud3.transformer.encoder.layers.1.self_attn.in_proj_weight", "bat_low.im_ex_boud3.transformer.encoder.layers.1.self_attn.in_proj_bias", "bat_low.im_ex_boud3.transformer.encoder.layers.1.self_attn.out_proj.weight", "bat_low.im_ex_boud3.transformer.encoder.layers.1.self_attn.out_proj.bias", "bat_low.im_ex_boud3.transformer.encoder.layers.1.linear1.weight", "bat_low.im_ex_boud3.transformer.encoder.layers.1.linear1.bias", "bat_low.im_ex_boud3.transformer.encoder.layers.1.linear2.weight", "bat_low.im_ex_boud3.transformer.encoder.layers.1.linear2.bias", "bat_low.im_ex_boud3.transformer.encoder.layers.1.norm1.weight", "bat_low.im_ex_boud3.transformer.encoder.layers.1.norm1.bias", "bat_low.im_ex_boud3.transformer.encoder.layers.1.norm2.weight", "bat_low.im_ex_boud3.transformer.encoder.layers.1.norm2.bias", "bat_low.im_ex_boud3.transformer.encoder.layers.1.BAG.0.weight", "bat_low.im_ex_boud3.transformer.encoder.layers.1.BAG.1.weight", "bat_low.im_ex_boud3.transformer.encoder.layers.1.BAG.1.bias", "bat_low.im_ex_boud3.transformer.encoder.layers.1.BAG.1.running_mean", "bat_low.im_ex_boud3.transformer.encoder.layers.1.BAG.1.running_var", "bat_low.im_ex_boud3.transformer.encoder.layers.1.BAG.3.weight", "bat_low.im_ex_boud3.transformer.encoder.layers.1.BAG.4.weight", "bat_low.im_ex_boud3.transformer.encoder.layers.1.BAG.4.bias", "bat_low.im_ex_boud3.transformer.encoder.layers.1.BAG.4.running_mean", "bat_low.im_ex_boud3.transformer.encoder.layers.1.BAG.4.running_var", "bat_low.im_ex_boud3.transformer.encoder.layers.1.BAG.6.weight", "bat_low.im_ex_boud3.transformer.encoder.layers.1.BAG.6.bias", "bat_low.im_ex_boud3.transformer.encoder.norm.weight", "bat_low.im_ex_boud3.transformer.encoder.norm.bias", "bat_low.im_ex_boud3.transformer.decoder.layers.0.self_attn.in_proj_weight", "bat_low.im_ex_boud3.transformer.decoder.layers.0.self_attn.in_proj_bias", "bat_low.im_ex_boud3.transformer.decoder.layers.0.self_attn.out_proj.weight", "bat_low.im_ex_boud3.transformer.decoder.layers.0.self_attn.out_proj.bias", "bat_low.im_ex_boud3.transformer.decoder.layers.0.multihead_attn.in_proj_weight", "bat_low.im_ex_boud3.transformer.decoder.layers.0.multihead_attn.in_proj_bias", "bat_low.im_ex_boud3.transformer.decoder.layers.0.multihead_attn.out_proj.weight", "bat_low.im_ex_boud3.transformer.decoder.layers.0.multihead_attn.out_proj.bias", "bat_low.im_ex_boud3.transformer.decoder.layers.0.linear1.weight", "bat_low.im_ex_boud3.transformer.decoder.layers.0.linear1.bias", "bat_low.im_ex_boud3.transformer.decoder.layers.0.linear2.weight", "bat_low.im_ex_boud3.transformer.decoder.layers.0.linear2.bias", "bat_low.im_ex_boud3.transformer.decoder.layers.0.norm1.weight", "bat_low.im_ex_boud3.transformer.decoder.layers.0.norm1.bias", "bat_low.im_ex_boud3.transformer.decoder.layers.0.norm2.weight", "bat_low.im_ex_boud3.transformer.decoder.layers.0.norm2.bias", "bat_low.im_ex_boud3.transformer.decoder.layers.0.norm3.weight", "bat_low.im_ex_boud3.transformer.decoder.layers.0.norm3.bias", "bat_low.im_ex_boud3.transformer.decoder.layers.1.self_attn.in_proj_weight", "bat_low.im_ex_boud3.transformer.decoder.layers.1.self_attn.in_proj_bias", "bat_low.im_ex_boud3.transformer.decoder.layers.1.self_attn.out_proj.weight", "bat_low.im_ex_boud3.transformer.decoder.layers.1.self_attn.out_proj.bias", "bat_low.im_ex_boud3.transformer.decoder.layers.1.multihead_attn.in_proj_weight", "bat_low.im_ex_boud3.transformer.decoder.layers.1.multihead_attn.in_proj_bias", "bat_low.im_ex_boud3.transformer.decoder.layers.1.multihead_attn.out_proj.weight", "bat_low.im_ex_boud3.transformer.decoder.layers.1.multihead_attn.out_proj.bias", "bat_low.im_ex_boud3.transformer.decoder.layers.1.linear1.weight", "bat_low.im_ex_boud3.transformer.decoder.layers.1.linear1.bias", "bat_low.im_ex_boud3.transformer.decoder.layers.1.linear2.weight", "bat_low.im_ex_boud3.transformer.decoder.layers.1.linear2.bias", "bat_low.im_ex_boud3.transformer.decoder.layers.1.norm1.weight", "bat_low.im_ex_boud3.transformer.decoder.layers.1.norm1.bias", "bat_low.im_ex_boud3.transformer.decoder.layers.1.norm2.weight", "bat_low.im_ex_boud3.transformer.decoder.layers.1.norm2.bias", "bat_low.im_ex_boud3.transformer.decoder.layers.1.norm3.weight", "bat_low.im_ex_boud3.transformer.decoder.layers.1.norm3.bias", "bat_low.im_ex_boud3.transformer.decoder.norm.weight", "bat_low.im_ex_boud3.transformer.decoder.norm.bias", "bat_low.cross_attention_3_1.xbl.cross_attn.in_proj_weight", "bat_low.cross_attention_3_1.xbl.cross_attn.in_proj_bias", "bat_low.cross_attention_3_1.xbl.cross_attn.out_proj.weight", "bat_low.cross_attention_3_1.xbl.cross_attn.out_proj.bias", "bat_low.cross_attention_3_1.xbl.linear1.weight", "bat_low.cross_attention_3_1.xbl.linear1.bias", "bat_low.cross_attention_3_1.xbl.linear2.weight", "bat_low.cross_attention_3_1.xbl.linear2.bias", "bat_low.cross_attention_3_1.xbl.norm1.weight", "bat_low.cross_attention_3_1.xbl.norm1.bias", "bat_low.cross_attention_3_1.xbl.norm2.weight", "bat_low.cross_attention_3_1.xbl.norm2.bias", "bat_low.cross_attention_3_1.xbl1.cross_attn.in_proj_weight", "bat_low.cross_attention_3_1.xbl1.cross_attn.in_proj_bias", "bat_low.cross_attention_3_1.xbl1.cross_attn.out_proj.weight", "bat_low.cross_attention_3_1.xbl1.cross_attn.out_proj.bias", "bat_low.cross_attention_3_1.xbl1.linear1.weight", "bat_low.cross_attention_3_1.xbl1.linear1.bias", "bat_low.cross_attention_3_1.xbl1.linear2.weight", "bat_low.cross_attention_3_1.xbl1.linear2.bias", "bat_low.cross_attention_3_1.xbl1.norm1.weight", "bat_low.cross_attention_3_1.xbl1.norm1.bias", "bat_low.cross_attention_3_1.xbl1.norm2.weight", "bat_low.cross_attention_3_1.xbl1.norm2.bias", "bat_low.cross_attention_3_1.mlp.weight", "bat_low.cross_attention_3_1.mlp.bias", "bat_low.cross_attention_3_2.xbl.cross_attn.in_proj_weight", "bat_low.cross_attention_3_2.xbl.cross_attn.in_proj_bias", "bat_low.cross_attention_3_2.xbl.cross_attn.out_proj.weight", "bat_low.cross_attention_3_2.xbl.cross_attn.out_proj.bias", "bat_low.cross_attention_3_2.xbl.linear1.weight", "bat_low.cross_attention_3_2.xbl.linear1.bias", "bat_low.cross_attention_3_2.xbl.linear2.weight", "bat_low.cross_attention_3_2.xbl.linear2.bias", "bat_low.cross_attention_3_2.xbl.norm1.weight", "bat_low.cross_attention_3_2.xbl.norm1.bias", "bat_low.cross_attention_3_2.xbl.norm2.weight", "bat_low.cross_attention_3_2.xbl.norm2.bias", "bat_low.cross_attention_3_2.xbl1.cross_attn.in_proj_weight", "bat_low.cross_attention_3_2.xbl1.cross_attn.in_proj_bias", "bat_low.cross_attention_3_2.xbl1.cross_attn.out_proj.weight", "bat_low.cross_attention_3_2.xbl1.cross_attn.out_proj.bias", "bat_low.cross_attention_3_2.xbl1.linear1.weight", "bat_low.cross_attention_3_2.xbl1.linear1.bias", "bat_low.cross_attention_3_2.xbl1.linear2.weight", "bat_low.cross_attention_3_2.xbl1.linear2.bias", "bat_low.cross_attention_3_2.xbl1.norm1.weight", "bat_low.cross_attention_3_2.xbl1.norm1.bias", "bat_low.cross_attention_3_2.xbl1.norm2.weight", "bat_low.cross_attention_3_2.xbl1.norm2.bias", "bat_low.cross_attention_3_2.mlp.weight", "bat_low.cross_attention_3_2.mlp.bias", "bat_low.trans_out_conv.weight", "bat_low.trans_out_conv.bias". Unexpected key(s) in state_dict: "query_positions", "row_embedding", "column_embedding", "deeplab.resnet.resnet.0.weight", "deeplab.resnet.resnet.1.weight", "deeplab.resnet.resnet.1.bias", "deeplab.resnet.resnet.1.running_mean", "deeplab.resnet.resnet.1.running_var", "deeplab.resnet.resnet.1.num_batches_tracked", "deeplab.resnet.resnet.4.0.conv1.weight", "deeplab.resnet.resnet.4.0.bn1.weight", "deeplab.resnet.resnet.4.0.bn1.bias", "deeplab.resnet.resnet.4.0.bn1.running_mean", "deeplab.resnet.resnet.4.0.bn1.running_var", "deeplab.resnet.resnet.4.0.bn1.num_batches_tracked", "deeplab.resnet.resnet.4.0.conv2.weight", "deeplab.resnet.resnet.4.0.bn2.weight", "deeplab.resnet.resnet.4.0.bn2.bias", "deeplab.resnet.resnet.4.0.bn2.running_mean", "deeplab.resnet.resnet.4.0.bn2.running_var", "deeplab.resnet.resnet.4.0.bn2.num_batches_tracked", "deeplab.resnet.resnet.4.0.conv3.weight", "deeplab.resnet.resnet.4.0.bn3.weight", "deeplab.resnet.resnet.4.0.bn3.bias", "deeplab.resnet.resnet.4.0.bn3.running_mean", "deeplab.resnet.resnet.4.0.bn3.running_var", "deeplab.resnet.resnet.4.0.bn3.num_batches_tracked", "deeplab.resnet.resnet.4.0.downsample.0.weight", "deeplab.resnet.resnet.4.0.downsample.1.weight", "deeplab.resnet.resnet.4.0.downsample.1.bias", "deeplab.resnet.resnet.4.0.downsample.1.running_mean", "deeplab.resnet.resnet.4.0.downsample.1.running_var", "deeplab.resnet.resnet.4.0.downsample.1.num_batches_tracked", "deeplab.resnet.resnet.4.1.conv1.weight", "deeplab.resnet.resnet.4.1.bn1.weight", "deeplab.resnet.resnet.4.1.bn1.bias", "deeplab.resnet.resnet.4.1.bn1.running_mean", "deeplab.resnet.resnet.4.1.bn1.running_var", "deeplab.resnet.resnet.4.1.bn1.num_batches_tracked", "deeplab.resnet.resnet.4.1.conv2.weight", "deeplab.resnet.resnet.4.1.bn2.weight", "deeplab.resnet.resnet.4.1.bn2.bias", "deeplab.resnet.resnet.4.1.bn2.running_mean", "deeplab.resnet.resnet.4.1.bn2.running_var", "deeplab.resnet.resnet.4.1.bn2.num_batches_tracked", "deeplab.resnet.resnet.4.1.conv3.weight", "deeplab.resnet.resnet.4.1.bn3.weight", "deeplab.resnet.resnet.4.1.bn3.bias", "deeplab.resnet.resnet.4.1.bn3.running_mean", "deeplab.resnet.resnet.4.1.bn3.running_var", "deeplab.resnet.resnet.4.1.bn3.num_batches_tracked", "deeplab.resnet.resnet.4.2.conv1.weight", "deeplab.resnet.resnet.4.2.bn1.weight", "deeplab.resnet.resnet.4.2.bn1.bias", "deeplab.resnet.resnet.4.2.bn1.running_mean", "deeplab.resnet.resnet.4.2.bn1.running_var", "deeplab.resnet.resnet.4.2.bn1.num_batches_tracked", "deeplab.resnet.resnet.4.2.conv2.weight", "deeplab.resnet.resnet.4.2.bn2.weight", "deeplab.resnet.resnet.4.2.bn2.bias", "deeplab.resnet.resnet.4.2.bn2.running_mean", "deeplab.resnet.resnet.4.2.bn2.running_var", "deeplab.resnet.resnet.4.2.bn2.num_batches_tracked", "deeplab.resnet.resnet.4.2.conv3.weight", "deeplab.resnet.resnet.4.2.bn3.weight", "deeplab.resnet.resnet.4.2.bn3.bias", "deeplab.resnet.resnet.4.2.bn3.running_mean", "deeplab.resnet.resnet.4.2.bn3.running_var", "deeplab.resnet.resnet.4.2.bn3.num_batches_tracked", "deeplab.resnet.resnet.5.0.conv1.weight", "deeplab.resnet.resnet.5.0.bn1.weight", "deeplab.resnet.resnet.5.0.bn1.bias", "deeplab.resnet.resnet.5.0.bn1.running_mean", "deeplab.resnet.resnet.5.0.bn1.running_var", "deeplab.resnet.resnet.5.0.bn1.num_batches_tracked", "deeplab.resnet.resnet.5.0.conv2.weight", "deeplab.resnet.resnet.5.0.bn2.weight", "deeplab.resnet.resnet.5.0.bn2.bias", "deeplab.resnet.resnet.5.0.bn2.running_mean", "deeplab.resnet.resnet.5.0.bn2.running_var", "deeplab.resnet.resnet.5.0.bn2.num_batches_tracked", "deeplab.resnet.resnet.5.0.conv3.weight", "deeplab.resnet.resnet.5.0.bn3.weight", "deeplab.resnet.resnet.5.0.bn3.bias", "deeplab.resnet.resnet.5.0.bn3.running_mean", "deeplab.resnet.resnet.5.0.bn3.running_var", "deeplab.resnet.resnet.5.0.bn3.num_batches_tracked", "deeplab.resnet.resnet.5.0.downsample.0.weight", "deeplab.resnet.resnet.5.0.downsample.1.weight", "deeplab.resnet.resnet.5.0.downsample.1.bias", "deeplab.resnet.resnet.5.0.downsample.1.running_mean", "deeplab.resnet.resnet.5.0.downsample.1.running_var", "deeplab.resnet.resnet.5.0.downsample.1.num_batches_tracked", "deeplab.resnet.resnet.5.1.conv1.weight", "deeplab.resnet.resnet.5.1.bn1.weight", "deeplab.resnet.resnet.5.1.bn1.bias", "deeplab.resnet.resnet.5.1.bn1.running_mean", "deeplab.resnet.resnet.5.1.bn1.running_var", "deeplab.resnet.resnet.5.1.bn1.num_batches_tracked", "deeplab.resnet.resnet.5.1.conv2.weight", "deeplab.resnet.resnet.5.1.bn2.weight", "deeplab.resnet.resnet.5.1.bn2.bias", "deeplab.resnet.resnet.5.1.bn2.running_mean", "deeplab.resnet.resnet.5.1.bn2.running_var", "deeplab.resnet.resnet.5.1.bn2.num_batches_tracked", "deeplab.resnet.resnet.5.1.conv3.weight", "deeplab.resnet.resnet.5.1.bn3.weight", "deeplab.resnet.resnet.5.1.bn3.bias", "deeplab.resnet.resnet.5.1.bn3.running_mean", "deeplab.resnet.resnet.5.1.bn3.running_var", "deeplab.resnet.resnet.5.1.bn3.num_batches_tracked", "deeplab.resnet.resnet.5.2.conv1.weight", "deeplab.resnet.resnet.5.2.bn1.weight", "deeplab.resnet.resnet.5.2.bn1.bias", "deeplab.resnet.resnet.5.2.bn1.running_mean", "deeplab.resnet.resnet.5.2.bn1.running_var", "deeplab.resnet.resnet.5.2.bn1.num_batches_tracked", "deeplab.resnet.resnet.5.2.conv2.weight", "deeplab.resnet.resnet.5.2.bn2.weight", "deeplab.resnet.resnet.5.2.bn2.bias", "deeplab.resnet.resnet.5.2.bn2.running_mean", "deeplab.resnet.resnet.5.2.bn2.running_var", "deeplab.resnet.resnet.5.2.bn2.num_batches_tracked", "deeplab.resnet.resnet.5.2.conv3.weight", "deeplab.resnet.resnet.5.2.bn3.weight", "deeplab.resnet.resnet.5.2.bn3.bias", "deeplab.resnet.resnet.5.2.bn3.running_mean", "deeplab.resnet.resnet.5.2.bn3.running_var", "deeplab.resnet.resnet.5.2.bn3.num_batches_tracked", "deeplab.resnet.resnet.5.3.conv1.weight", "deeplab.resnet.resnet.5.3.bn1.weight", "deeplab.resnet.resnet.5.3.bn1.bias", "deeplab.resnet.resnet.5.3.bn1.running_mean", "deeplab.resnet.resnet.5.3.bn1.running_var", "deeplab.resnet.resnet.5.3.bn1.num_batches_tracked", "deeplab.resnet.resnet.5.3.conv2.weight", "deeplab.resnet.resnet.5.3.bn2.weight", "deeplab.resnet.resnet.5.3.bn2.bias", "deeplab.resnet.resnet.5.3.bn2.running_mean", "deeplab.resnet.resnet.5.3.bn2.running_var", "deeplab.resnet.resnet.5.3.bn2.num_batches_tracked", "deeplab.resnet.resnet.5.3.conv3.weight", "deeplab.resnet.resnet.5.3.bn3.weight", "deeplab.resnet.resnet.5.3.bn3.bias", "deeplab.resnet.resnet.5.3.bn3.running_mean", "deeplab.resnet.resnet.5.3.bn3.running_var", "deeplab.resnet.resnet.5.3.bn3.num_batches_tracked", "deeplab.resnet.resnet.6.0.conv1.weight", "deeplab.resnet.resnet.6.0.bn1.weight", "deeplab.resnet.resnet.6.0.bn1.bias", "deeplab.resnet.resnet.6.0.bn1.running_mean", "deeplab.resnet.resnet.6.0.bn1.running_var", "deeplab.resnet.resnet.6.0.bn1.num_batches_tracked", "deeplab.resnet.resnet.6.0.conv2.weight", "deeplab.resnet.resnet.6.0.bn2.weight", "deeplab.resnet.resnet.6.0.bn2.bias", "deeplab.resnet.resnet.6.0.bn2.running_mean", "deeplab.resnet.resnet.6.0.bn2.running_var", "deeplab.resnet.resnet.6.0.bn2.num_batches_tracked", "deeplab.resnet.resnet.6.0.conv3.weight", "deeplab.resnet.resnet.6.0.bn3.weight", "deeplab.resnet.resnet.6.0.bn3.bias", "deeplab.resnet.resnet.6.0.bn3.running_mean", "deeplab.resnet.resnet.6.0.bn3.running_var", "deeplab.resnet.resnet.6.0.bn3.num_batches_tracked", "deeplab.resnet.resnet.6.0.downsample.0.weight", "deeplab.resnet.resnet.6.0.downsample.1.weight", "deeplab.resnet.resnet.6.0.downsample.1.bias", "deeplab.resnet.resnet.6.0.downsample.1.running_mean", "deeplab.resnet.resnet.6.0.downsample.1.running_var", "deeplab.resnet.resnet.6.0.downsample.1.num_batches_tracked", "deeplab.resnet.resnet.6.1.conv1.weight", "deeplab.resnet.resnet.6.1.bn1.weight", "deeplab.resnet.resnet.6.1.bn1.bias", "deeplab.resnet.resnet.6.1.bn1.running_mean", "deeplab.resnet.resnet.6.1.bn1.running_var", "deeplab.resnet.resnet.6.1.bn1.num_batches_tracked", "deeplab.resnet.resnet.6.1.conv2.weight", "deeplab.resnet.resnet.6.1.bn2.weight", "deeplab.resnet.resnet.6.1.bn2.bias", "deeplab.resnet.resnet.6.1.bn2.running_mean", "deeplab.resnet.resnet.6.1.bn2.running_var", "deeplab.resnet.resnet.6.1.bn2.num_batches_tracked", "deeplab.resnet.resnet.6.1.conv3.weight", "deeplab.resnet.resnet.6.1.bn3.weight", "deeplab.resnet.resnet.6.1.bn3.bias", "deeplab.resnet.resnet.6.1.bn3.running_mean", "deeplab.resnet.resnet.6.1.bn3.running_var", "deeplab.resnet.resnet.6.1.bn3.num_batches_tracked", "deeplab.resnet.resnet.6.2.conv1.weight", "deeplab.resnet.resnet.6.2.bn1.weight", "deeplab.resnet.resnet.6.2.bn1.bias", "deeplab.resnet.resnet.6.2.bn1.running_mean", "deeplab.resnet.resnet.6.2.bn1.running_var", "deeplab.resnet.resnet.6.2.bn1.num_batches_tracked", "deeplab.resnet.resnet.6.2.conv2.weight", "deeplab.resnet.resnet.6.2.bn2.weight", "deeplab.resnet.resnet.6.2.bn2.bias", "deeplab.resnet.resnet.6.2.bn2.running_mean", "deeplab.resnet.resnet.6.2.bn2.running_var", "deeplab.resnet.resnet.6.2.bn2.num_batches_tracked", "deeplab.resnet.resnet.6.2.conv3.weight", "deeplab.resnet.resnet.6.2.bn3.weight", "deeplab.resnet.resnet.6.2.bn3.bias", "deeplab.resnet.resnet.6.2.bn3.running_mean", "deeplab.resnet.resnet.6.2.bn3.running_var", "deeplab.resnet.resnet.6.2.bn3.num_batches_tracked", "deeplab.resnet.resnet.6.3.conv1.weight", "deeplab.resnet.resnet.6.3.bn1.weight", "deeplab.resnet.resnet.6.3.bn1.bias", "deeplab.resnet.resnet.6.3.bn1.running_mean", "deeplab.resnet.resnet.6.3.bn1.running_var", "deeplab.resnet.resnet.6.3.bn1.num_batches_tracked", "deeplab.resnet.resnet.6.3.conv2.weight", "deeplab.resnet.resnet.6.3.bn2.weight", "deeplab.resnet.resnet.6.3.bn2.bias", "deeplab.resnet.resnet.6.3.bn2.running_mean", "deeplab.resnet.resnet.6.3.bn2.running_var", "deeplab.resnet.resnet.6.3.bn2.num_batches_tracked", "deeplab.resnet.resnet.6.3.conv3.weight", "deeplab.resnet.resnet.6.3.bn3.weight", "deeplab.resnet.resnet.6.3.bn3.bias", "deeplab.resnet.resnet.6.3.bn3.running_mean", "deeplab.resnet.resnet.6.3.bn3.running_var", "deeplab.resnet.resnet.6.3.bn3.num_batches_tracked", "deeplab.resnet.resnet.6.4.conv1.weight", "deeplab.resnet.resnet.6.4.bn1.weight", "deeplab.resnet.resnet.6.4.bn1.bias", "deeplab.resnet.resnet.6.4.bn1.running_mean", "deeplab.resnet.resnet.6.4.bn1.running_var", "deeplab.resnet.resnet.6.4.bn1.num_batches_tracked", "deeplab.resnet.resnet.6.4.conv2.weight", "deeplab.resnet.resnet.6.4.bn2.weight", "deeplab.resnet.resnet.6.4.bn2.bias", "deeplab.resnet.resnet.6.4.bn2.running_mean", "deeplab.resnet.resnet.6.4.bn2.running_var", "deeplab.resnet.resnet.6.4.bn2.num_batches_tracked", "deeplab.resnet.resnet.6.4.conv3.weight", "deeplab.resnet.resnet.6.4.bn3.weight", "deeplab.resnet.resnet.6.4.bn3.bias", "deeplab.resnet.resnet.6.4.bn3.running_mean", "deeplab.resnet.resnet.6.4.bn3.running_var", "deeplab.resnet.resnet.6.4.bn3.num_batches_tracked", "deeplab.resnet.resnet.6.5.conv1.weight", "deeplab.resnet.resnet.6.5.bn1.weight", "deeplab.resnet.resnet.6.5.bn1.bias", "deeplab.resnet.resnet.6.5.bn1.running_mean", "deeplab.resnet.resnet.6.5.bn1.running_var", "deeplab.resnet.resnet.6.5.bn1.num_batches_tracked", "deeplab.resnet.resnet.6.5.conv2.weight", "deeplab.resnet.resnet.6.5.bn2.weight", "deeplab.resnet.resnet.6.5.bn2.bias", "deeplab.resnet.resnet.6.5.bn2.running_mean", "deeplab.resnet.resnet.6.5.bn2.running_var", "deeplab.resnet.resnet.6.5.bn2.num_batches_tracked", "deeplab.resnet.resnet.6.5.conv3.weight", "deeplab.resnet.resnet.6.5.bn3.weight", "deeplab.resnet.resnet.6.5.bn3.bias", "deeplab.resnet.resnet.6.5.bn3.running_mean", "deeplab.resnet.resnet.6.5.bn3.running_var", "deeplab.resnet.resnet.6.5.bn3.num_batches_tracked", "deeplab.resnet.layer5.0.conv1.weight", "deeplab.resnet.layer5.0.bn1.weight", "deeplab.resnet.layer5.0.bn1.bias", "deeplab.resnet.layer5.0.bn1.running_mean", "deeplab.resnet.layer5.0.bn1.running_var", "deeplab.resnet.layer5.0.bn1.num_batches_tracked", "deeplab.resnet.layer5.0.conv2.weight", "deeplab.resnet.layer5.0.bn2.weight", "deeplab.resnet.layer5.0.bn2.bias", "deeplab.resnet.layer5.0.bn2.running_mean", "deeplab.resnet.layer5.0.bn2.running_var", "deeplab.resnet.layer5.0.bn2.num_batches_tracked", "deeplab.resnet.layer5.0.conv3.weight", "deeplab.resnet.layer5.0.bn3.weight", "deeplab.resnet.layer5.0.bn3.bias", "deeplab.resnet.layer5.0.bn3.running_mean", "deeplab.resnet.layer5.0.bn3.running_var", "deeplab.resnet.layer5.0.bn3.num_batches_tracked", "deeplab.resnet.layer5.0.downsample.0.weight", "deeplab.resnet.layer5.0.downsample.1.weight", "deeplab.resnet.layer5.0.downsample.1.bias", "deeplab.resnet.layer5.0.downsample.1.running_mean", "deeplab.resnet.layer5.0.downsample.1.running_var", "deeplab.resnet.layer5.0.downsample.1.num_batches_tracked", "deeplab.resnet.layer5.1.conv1.weight", "deeplab.resnet.layer5.1.bn1.weight", "deeplab.resnet.layer5.1.bn1.bias", "deeplab.resnet.layer5.1.bn1.running_mean", "deeplab.resnet.layer5.1.bn1.running_var", "deeplab.resnet.layer5.1.bn1.num_batches_tracked", "deeplab.resnet.layer5.1.conv2.weight", "deeplab.resnet.layer5.1.bn2.weight", "deeplab.resnet.layer5.1.bn2.bias", "deeplab.resnet.layer5.1.bn2.running_mean", "deeplab.resnet.layer5.1.bn2.running_var", "deeplab.resnet.layer5.1.bn2.num_batches_tracked", "deeplab.resnet.layer5.1.conv3.weight", "deeplab.resnet.layer5.1.bn3.weight", "deeplab.resnet.layer5.1.bn3.bias", "deeplab.resnet.layer5.1.bn3.running_mean", "deeplab.resnet.layer5.1.bn3.running_var", "deeplab.resnet.layer5.1.bn3.num_batches_tracked", "deeplab.resnet.layer5.2.conv1.weight", "deeplab.resnet.layer5.2.bn1.weight", "deeplab.resnet.layer5.2.bn1.bias", "deeplab.resnet.layer5.2.bn1.running_mean", "deeplab.resnet.layer5.2.bn1.running_var", "deeplab.resnet.layer5.2.bn1.num_batches_tracked", "deeplab.resnet.layer5.2.conv2.weight", "deeplab.resnet.layer5.2.bn2.weight", "deeplab.resnet.layer5.2.bn2.bias", "deeplab.resnet.layer5.2.bn2.running_mean", "deeplab.resnet.layer5.2.bn2.running_var", "deeplab.resnet.layer5.2.bn2.num_batches_tracked", "deeplab.resnet.layer5.2.conv3.weight", "deeplab.resnet.layer5.2.bn3.weight", "deeplab.resnet.layer5.2.bn3.bias", "deeplab.resnet.layer5.2.bn3.running_mean", "deeplab.resnet.layer5.2.bn3.running_var", "deeplab.resnet.layer5.2.bn3.num_batches_tracked", "deeplab.aspp.conv_1x1_1.weight", "deeplab.aspp.conv_1x1_1.bias", "deeplab.aspp.bn_conv_1x1_1.weight", "deeplab.aspp.bn_conv_1x1_1.bias", "deeplab.aspp.bn_conv_1x1_1.running_mean", "deeplab.aspp.bn_conv_1x1_1.running_var", "deeplab.aspp.bn_conv_1x1_1.num_batches_tracked", "deeplab.aspp.conv_3x3_1.weight", "deeplab.aspp.conv_3x3_1.bias", "deeplab.aspp.bn_conv_3x3_1.weight", "deeplab.aspp.bn_conv_3x3_1.bias", "deeplab.aspp.bn_conv_3x3_1.running_mean", "deeplab.aspp.bn_conv_3x3_1.running_var", "deeplab.aspp.bn_conv_3x3_1.num_batches_tracked", "deeplab.aspp.conv_3x3_2.weight", "deeplab.aspp.conv_3x3_2.bias", "deeplab.aspp.bn_conv_3x3_2.weight", "deeplab.aspp.bn_conv_3x3_2.bias", "deeplab.aspp.bn_conv_3x3_2.running_mean", "deeplab.aspp.bn_conv_3x3_2.running_var", "deeplab.aspp.bn_conv_3x3_2.num_batches_tracked", "deeplab.aspp.conv_3x3_3.weight", "deeplab.aspp.conv_3x3_3.bias", "deeplab.aspp.bn_conv_3x3_3.weight", "deeplab.aspp.bn_conv_3x3_3.bias", "deeplab.aspp.bn_conv_3x3_3.running_mean", "deeplab.aspp.bn_conv_3x3_3.running_var", "deeplab.aspp.bn_conv_3x3_3.num_batches_tracked", "deeplab.aspp.conv_1x1_2.weight", "deeplab.aspp.conv_1x1_2.bias", "deeplab.aspp.bn_conv_1x1_2.weight", "deeplab.aspp.bn_conv_1x1_2.bias", "deeplab.aspp.bn_conv_1x1_2.running_mean", "deeplab.aspp.bn_conv_1x1_2.running_var", "deeplab.aspp.bn_conv_1x1_2.num_batches_tracked", "deeplab.aspp.conv_1x1_3.weight", "deeplab.aspp.conv_1x1_3.bias", "deeplab.aspp.bn_conv_1x1_3.weight", "deeplab.aspp.bn_conv_1x1_3.bias", "deeplab.aspp.bn_conv_1x1_3.running_mean", "deeplab.aspp.bn_conv_1x1_3.running_var", "deeplab.aspp.bn_conv_1x1_3.num_batches_tracked", "deeplab.aspp.conv_1x1_4.weight", "deeplab.aspp.conv_1x1_4.bias", "convolution_mapping.weight", "convolution_mapping.bias", "transformer.encoder.layers.0.self_attn.in_proj_weight", "transformer.encoder.layers.0.self_attn.in_proj_bias", "transformer.encoder.layers.0.self_attn.out_proj.weight", "transformer.encoder.layers.0.self_attn.out_proj.bias", "transformer.encoder.layers.0.linear1.weight", "transformer.encoder.layers.0.linear1.bias", "transformer.encoder.layers.0.linear2.weight", "transformer.encoder.layers.0.linear2.bias", "transformer.encoder.layers.0.norm1.weight", "transformer.encoder.layers.0.norm1.bias", "transformer.encoder.layers.0.norm2.weight", "transformer.encoder.layers.0.norm2.bias", "transformer.encoder.layers.0.BAG.0.weight", "transformer.encoder.layers.0.BAG.1.weight", "transformer.encoder.layers.0.BAG.1.bias", "transformer.encoder.layers.0.BAG.1.running_mean", "transformer.encoder.layers.0.BAG.1.running_var", "transformer.encoder.layers.0.BAG.1.num_batches_tracked", "transformer.encoder.layers.0.BAG.3.weight", "transformer.encoder.layers.0.BAG.4.weight", "transformer.encoder.layers.0.BAG.4.bias", "transformer.encoder.layers.0.BAG.4.running_mean", "transformer.encoder.layers.0.BAG.4.running_var", "transformer.encoder.layers.0.BAG.4.num_batches_tracked", "transformer.encoder.layers.0.BAG.6.weight", "transformer.encoder.layers.0.BAG.6.bias", "transformer.encoder.layers.1.self_attn.in_proj_weight", "transformer.encoder.layers.1.self_attn.in_proj_bias", "transformer.encoder.layers.1.self_attn.out_proj.weight", "transformer.encoder.layers.1.self_attn.out_proj.bias", "transformer.encoder.layers.1.linear1.weight", "transformer.encoder.layers.1.linear1.bias", "transformer.encoder.layers.1.linear2.weight", "transformer.encoder.layers.1.linear2.bias", "transformer.encoder.layers.1.norm1.weight", "transformer.encoder.layers.1.norm1.bias", "transformer.encoder.layers.1.norm2.weight", "transformer.encoder.layers.1.norm2.bias", "transformer.encoder.layers.1.BAG.0.weight", "transformer.encoder.layers.1.BAG.1.weight", "transformer.encoder.layers.1.BAG.1.bias", "transformer.encoder.layers.1.BAG.1.running_mean", "transformer.encoder.layers.1.BAG.1.running_var", "transformer.encoder.layers.1.BAG.1.num_batches_tracked", "transformer.encoder.layers.1.BAG.3.weight", "transformer.encoder.layers.1.BAG.4.weight", "transformer.encoder.layers.1.BAG.4.bias", "transformer.encoder.layers.1.BAG.4.running_mean", "transformer.encoder.layers.1.BAG.4.running_var", "transformer.encoder.layers.1.BAG.4.num_batches_tracked", "transformer.encoder.layers.1.BAG.6.weight", "transformer.encoder.layers.1.BAG.6.bias", "transformer.encoder.layers.2.self_attn.in_proj_weight", "transformer.encoder.layers.2.self_attn.in_proj_bias", "transformer.encoder.layers.2.self_attn.out_proj.weight", "transformer.encoder.layers.2.self_attn.out_proj.bias", "transformer.encoder.layers.2.linear1.weight", "transformer.encoder.layers.2.linear1.bias", "transformer.encoder.layers.2.linear2.weight", "transformer.encoder.layers.2.linear2.bias", "transformer.encoder.layers.2.norm1.weight", "transformer.encoder.layers.2.norm1.bias", "transformer.encoder.layers.2.norm2.weight", "transformer.encoder.layers.2.norm2.bias", "transformer.encoder.layers.2.BAG.0.weight", "transformer.encoder.layers.2.BAG.1.weight", "transformer.encoder.layers.2.BAG.1.bias", "transformer.encoder.layers.2.BAG.1.running_mean", "transformer.encoder.layers.2.BAG.1.running_var", "transformer.encoder.layers.2.BAG.1.num_batches_tracked", "transformer.encoder.layers.2.BAG.3.weight", "transformer.encoder.layers.2.BAG.4.weight", "transformer.encoder.layers.2.BAG.4.bias", "transformer.encoder.layers.2.BAG.4.running_mean", "transformer.encoder.layers.2.BAG.4.running_var", "transformer.encoder.layers.2.BAG.4.num_batches_tracked", "transformer.encoder.layers.2.BAG.6.weight", "transformer.encoder.layers.2.BAG.6.bias", "transformer.encoder.layers.3.self_attn.in_proj_weight", "transformer.encoder.layers.3.self_attn.in_proj_bias", "transformer.encoder.layers.3.self_attn.out_proj.weight", "transformer.encoder.layers.3.self_attn.out_proj.bias", "transformer.encoder.layers.3.linear1.weight", "transformer.encoder.layers.3.linear1.bias", "transformer.encoder.layers.3.linear2.weight", "transformer.encoder.layers.3.linear2.bias", "transformer.encoder.layers.3.norm1.weight", "transformer.encoder.layers.3.norm1.bias", "transformer.encoder.layers.3.norm2.weight", "transformer.encoder.layers.3.norm2.bias", "transformer.encoder.layers.3.BAG.0.weight", "transformer.encoder.layers.3.BAG.1.weight", "transformer.encoder.layers.3.BAG.1.bias", "transformer.encoder.layers.3.BAG.1.running_mean", "transformer.encoder.layers.3.BAG.1.running_var", "transformer.encoder.layers.3.BAG.1.num_batches_tracked", "transformer.encoder.layers.3.BAG.3.weight", "transformer.encoder.layers.3.BAG.4.weight", "transformer.encoder.layers.3.BAG.4.bias", "transformer.encoder.layers.3.BAG.4.running_mean", "transformer.encoder.layers.3.BAG.4.running_var", "transformer.encoder.layers.3.BAG.4.num_batches_tracked", "transformer.encoder.layers.3.BAG.6.weight", "transformer.encoder.layers.3.BAG.6.bias", "transformer.encoder.layers.4.self_attn.in_proj_weight", "transformer.encoder.layers.4.self_attn.in_proj_bias", "transformer.encoder.layers.4.self_attn.out_proj.weight", "transformer.encoder.layers.4.self_attn.out_proj.bias", "transformer.encoder.layers.4.linear1.weight", "transformer.encoder.layers.4.linear1.bias", "transformer.encoder.layers.4.linear2.weight", "transformer.encoder.layers.4.linear2.bias", "transformer.encoder.layers.4.norm1.weight", "transformer.encoder.layers.4.norm1.bias", "transformer.encoder.layers.4.norm2.weight", "transformer.encoder.layers.4.norm2.bias", "transformer.encoder.layers.4.BAG.0.weight", "transformer.encoder.layers.4.BAG.1.weight", "transformer.encoder.layers.4.BAG.1.bias", "transformer.encoder.layers.4.BAG.1.running_mean", "transformer.encoder.layers.4.BAG.1.running_var", "transformer.encoder.layers.4.BAG.1.num_batches_tracked", "transformer.encoder.layers.4.BAG.3.weight", "transformer.encoder.layers.4.BAG.4.weight", "transformer.encoder.layers.4.BAG.4.bias", "transformer.encoder.layers.4.BAG.4.running_mean", "transformer.encoder.layers.4.BAG.4.running_var", "transformer.encoder.layers.4.BAG.4.num_batches_tracked", "transformer.encoder.layers.4.BAG.6.weight", "transformer.encoder.layers.4.BAG.6.bias", "transformer.encoder.layers.5.self_attn.in_proj_weight", "transformer.encoder.layers.5.self_attn.in_proj_bias", "transformer.encoder.layers.5.self_attn.out_proj.weight", "transformer.encoder.layers.5.self_attn.out_proj.bias", "transformer.encoder.layers.5.linear1.weight", "transformer.encoder.layers.5.linear1.bias", "transformer.encoder.layers.5.linear2.weight", "transformer.encoder.layers.5.linear2.bias", "transformer.encoder.layers.5.norm1.weight", "transformer.encoder.layers.5.norm1.bias", "transformer.encoder.layers.5.norm2.weight", "transformer.encoder.layers.5.norm2.bias", "transformer.encoder.layers.5.BAG.0.weight", "transformer.encoder.layers.5.BAG.1.weight", "transformer.encoder.layers.5.BAG.1.bias", "transformer.encoder.layers.5.BAG.1.running_mean", "transformer.encoder.layers.5.BAG.1.running_var", "transformer.encoder.layers.5.BAG.1.num_batches_tracked", "transformer.encoder.layers.5.BAG.3.weight", "transformer.encoder.layers.5.BAG.4.weight", "transformer.encoder.layers.5.BAG.4.bias", "transformer.encoder.layers.5.BAG.4.running_mean", "transformer.encoder.layers.5.BAG.4.running_var", "transformer.encoder.layers.5.BAG.4.num_batches_tracked", "transformer.encoder.layers.5.BAG.6.weight", "transformer.encoder.layers.5.BAG.6.bias", "transformer.decoder.layers.0.self_attn.in_proj_weight", "transformer.decoder.layers.0.self_attn.in_proj_bias", "transformer.decoder.layers.0.self_attn.out_proj.weight", "transformer.decoder.layers.0.self_attn.out_proj.bias", "transformer.decoder.layers.0.multihead_attn.in_proj_weight", "transformer.decoder.layers.0.multihead_attn.in_proj_bias", "transformer.decoder.layers.0.multihead_attn.out_proj.weight", "transformer.decoder.layers.0.multihead_attn.out_proj.bias", "transformer.decoder.layers.0.linear1.weight", "transformer.decoder.layers.0.linear1.bias", "transformer.decoder.layers.0.linear2.weight", "transformer.decoder.layers.0.linear2.bias", "transformer.decoder.layers.0.norm1.weight", "transformer.decoder.layers.0.norm1.bias", "transformer.decoder.layers.0.norm2.weight", "transformer.decoder.layers.0.norm2.bias", "transformer.decoder.layers.0.norm3.weight", "transformer.decoder.layers.0.norm3.bias", "transformer.decoder.layers.1.self_attn.in_proj_weight", "transformer.decoder.layers.1.self_attn.in_proj_bias", "transformer.decoder.layers.1.self_attn.out_proj.weight", "transformer.decoder.layers.1.self_attn.out_proj.bias", "transformer.decoder.layers.1.multihead_attn.in_proj_weight", "transformer.decoder.layers.1.multihead_attn.in_proj_bias", "transformer.decoder.layers.1.multihead_attn.out_proj.weight", "transformer.decoder.layers.1.multihead_attn.out_proj.bias", "transformer.decoder.layers.1.linear1.weight", "transformer.decoder.layers.1.linear1.bias", "transformer.decoder.layers.1.linear2.weight", "transformer.decoder.layers.1.linear2.bias", "transformer.decoder.layers.1.norm1.weight", "transformer.decoder.layers.1.norm1.bias", "transformer.decoder.layers.1.norm2.weight", "transformer.decoder.layers.1.norm2.bias", "transformer.decoder.layers.1.norm3.weight", "transformer.decoder.layers.1.norm3.bias", "transformer.decoder.norm.weight", "transformer.decoder.norm.bias", "BCA.cross_attn.in_proj_weight", "BCA.cross_attn.in_proj_bias", "BCA.cross_attn.out_proj.weight", "BCA.cross_attn.out_proj.bias", "BCA.linear1.weight", "BCA.linear1.bias", "BCA.linear2.weight", "BCA.linear2.bias", "BCA.norm1.weight", "BCA.norm1.bias", "BCA.norm2.weight", "BCA.norm2.bias", "BCA.BAG.convs.0.0.weight", "BCA.BAG.convs.0.1.weight", "BCA.BAG.convs.0.1.bias", "BCA.BAG.convs.0.1.running_mean", "BCA.BAG.convs.0.1.running_var", "BCA.BAG.convs.0.1.num_batches_tracked", "BCA.BAG.convs.1.0.weight", "BCA.BAG.convs.1.1.weight", "BCA.BAG.convs.1.1.bias", "BCA.BAG.convs.1.1.running_mean", "BCA.BAG.convs.1.1.running_var", "BCA.BAG.convs.1.1.num_batches_tracked", "BCA.BAG.convs.2.0.weight", "BCA.BAG.convs.2.1.weight", "BCA.BAG.convs.2.1.bias", "BCA.BAG.convs.2.1.running_mean", "BCA.BAG.convs.2.1.running_var", "BCA.BAG.convs.2.1.num_batches_tracked", "BCA.BAG.convs.3.0.weight", "BCA.BAG.convs.3.1.weight", "BCA.BAG.convs.3.1.bias", "BCA.BAG.convs.3.1.running_mean", "BCA.BAG.convs.3.1.running_var", "BCA.BAG.convs.3.1.num_batches_tracked", "BCA.BAG.convs.4.0.weight", "BCA.BAG.convs.4.1.weight", "BCA.BAG.convs.4.1.bias", "BCA.BAG.convs.4.1.running_mean", "BCA.BAG.convs.4.1.running_var", "BCA.BAG.convs.4.1.num_batches_tracked", "BCA.BAG.conv_out.weight", "trans_out_conv.weight", "trans_out_conv.bias".

UniverseFan commented 4 months ago

@houjunlin hello, is it solved.