tangjiapeng
commented
1 month ago I feel you are using the unconditional model to run rearrange model.
Open sankydesai opened 1 month ago
tangjiapeng
commented
1 month ago I feel you are using the unconditional model to run rearrange model.
Error(s) in loading state_dict for DiffusionSceneLayout_DDPM: Missing key(s) in state_dict: "diffusion.model.final_conv.weight", "diffusion.model.final_conv.bias", "fc_arrange_condition.0.weight", "fc_arrange_condition.2.weight". Unexpected key(s) in state_dict: "diffusion.model.objfeat_embedf.0.weight", "diffusion.model.objfeat_embedf.0.bias", "diffusion.model.objfeat_embedf.2.weight", "diffusion.model.objfeat_embedf.2.bias", "diffusion.model.objfeat_embedf.4.weight", "diffusion.model.objfeat_embedf.4.bias", "diffusion.model.class_embedf.0.weight", "diffusion.model.class_embedf.0.bias", "diffusion.model.class_embedf.2.weight", "diffusion.model.class_embedf.2.bias", "diffusion.model.class_embedf.4.weight", "diffusion.model.class_embedf.4.bias", "diffusion.model.bbox_embedf.0.weight", "diffusion.model.bbox_embedf.0.bias", "diffusion.model.bbox_embedf.2.weight", "diffusion.model.bbox_embedf.2.bias", "diffusion.model.bbox_embedf.4.weight", "diffusion.model.bbox_embedf.4.bias", "diffusion.model.objfeat_hidden2output.0.weight", "diffusion.model.objfeat_hidden2output.0.bias", "diffusion.model.objfeat_hidden2output.2.weight", "diffusion.model.objfeat_hidden2output.2.bias", "diffusion.model.objfeat_hidden2output.4.weight", "diffusion.model.objfeat_hidden2output.4.bias", "diffusion.model.class_hidden2output.0.weight", "diffusion.model.class_hidden2output.0.bias", "diffusion.model.class_hidden2output.2.weight", "diffusion.model.class_hidden2output.2.bias", "diffusion.model.class_hidden2output.4.weight", "diffusion.model.class_hidden2output.4.bias", "diffusion.model.bbox_hidden2output.0.weight", "diffusion.model.bbox_hidden2output.0.bias", "diffusion.model.bbox_hidden2output.2.weight", "diffusion.model.bbox_hidden2output.2.bias", "diffusion.model.bbox_hidden2output.4.weight", "diffusion.model.bbox_hidden2output.4.bias". size mismatch for diffusion.model.init_conv.weight: copying a param with shape torch.Size([512, 512, 1]) from checkpoint, the shape in current model is torch.Size([512, 5, 1]). size mismatch for diffusion.model.downs.0.0.mlp.1.weight: copying a param with shape torch.Size([1024, 128]) from checkpoint, the shape in current model is torch.Size([1024, 512]). size mismatch for diffusion.model.downs.1.0.mlp.1.weight: copying a param with shape torch.Size([1024, 128]) from checkpoint, the shape in current model is torch.Size([1024, 512]). size mismatch for diffusion.model.downs.2.0.mlp.1.weight: copying a param with shape torch.Size([1024, 128]) from checkpoint, the shape in current model is torch.Size([1024, 512]). size mismatch for diffusion.model.downs.3.0.mlp.1.weight: copying a param with shape torch.Size([1024, 128]) from checkpoint, the shape in current model is torch.Size([1024, 512]). size mismatch for diffusion.model.ups.0.0.mlp.1.weight: copying a param with shape torch.Size([1024, 128]) from checkpoint, the shape in current model is torch.Size([1024, 512]). size mismatch for diffusion.model.ups.1.0.mlp.1.weight: copying a param with shape torch.Size([1024, 128]) from checkpoint, the shape in current model is torch.Size([1024, 512]). size mismatch for diffusion.model.ups.2.0.mlp.1.weight: copying a param with shape torch.Size([1024, 128]) from checkpoint, the shape in current model is torch.Size([1024, 512]). size mismatch for diffusion.model.ups.3.0.mlp.1.weight: copying a param with shape torch.Size([1024, 128]) from checkpoint, the shape in current model is torch.Size([1024, 512]). size mismatch for diffusion.model.mid_block0.mlp.1.weight: copying a param with shape torch.Size([1024, 128]) from checkpoint, the shape in current model is torch.Size([1024, 512]). File "/home/ubuntu/.local/lib/python3.8/site-packages/torch/nn/modules/module.py", line 2189, in load_state_dict raise RuntimeError('Error(s) in loading state_dict for {}:\n\t{}'.format( File "/home/ubuntu/dev/diffuscene/DiffuScene/scene_synthesis/networks/init.py", line 64, in build_network network.load_state_dict( File "/home/ubuntu/dev/diffuscene/DiffuScene/scripts/completionrearrange.py", line 236, in main network, , _ = build_network( File "/home/ubuntu/dev/diffuscene/DiffuScene/scripts/completion_rearrange.py", line 542, in (Current frame)
main(sys.argv[1:])
File "/opt/conda/envs/diffuscene/lib/python3.8/runpy.py", line 87, in _run_code
exec(code, run_globals)
File "/opt/conda/envs/diffuscene/lib/python3.8/runpy.py", line 194, in _run_module_as_main
return _run_code(code, main_globals, None,
RuntimeError: Error(s) in loading state_dict for DiffusionSceneLayout_DDPM:
Missing key(s) in state_dict: "diffusion.model.final_conv.weight", "diffusion.model.final_conv.bias", "fc_arrange_condition.0.weight", "fc_arrange_condition.2.weight".
Unexpected key(s) in state_dict: "diffusion.model.objfeat_embedf.0.weight", "diffusion.model.objfeat_embedf.0.bias", "diffusion.model.objfeat_embedf.2.weight", "diffusion.model.objfeat_embedf.2.bias", "diffusion.model.objfeat_embedf.4.weight", "diffusion.model.objfeat_embedf.4.bias", "diffusion.model.class_embedf.0.weight", "diffusion.model.class_embedf.0.bias", "diffusion.model.class_embedf.2.weight", "diffusion.model.class_embedf.2.bias", "diffusion.model.class_embedf.4.weight", "diffusion.model.class_embedf.4.bias", "diffusion.model.bbox_embedf.0.weight", "diffusion.model.bbox_embedf.0.bias", "diffusion.model.bbox_embedf.2.weight", "diffusion.model.bbox_embedf.2.bias", "diffusion.model.bbox_embedf.4.weight", "diffusion.model.bbox_embedf.4.bias", "diffusion.model.objfeat_hidden2output.0.weight", "diffusion.model.objfeat_hidden2output.0.bias", "diffusion.model.objfeat_hidden2output.2.weight", "diffusion.model.objfeat_hidden2output.2.bias", "diffusion.model.objfeat_hidden2output.4.weight", "diffusion.model.objfeat_hidden2output.4.bias", "diffusion.model.class_hidden2output.0.weight", "diffusion.model.class_hidden2output.0.bias", "diffusion.model.class_hidden2output.2.weight", "diffusion.model.class_hidden2output.2.bias", "diffusion.model.class_hidden2output.4.weight", "diffusion.model.class_hidden2output.4.bias", "diffusion.model.bbox_hidden2output.0.weight", "diffusion.model.bbox_hidden2output.0.bias", "diffusion.model.bbox_hidden2output.2.weight", "diffusion.model.bbox_hidden2output.2.bias", "diffusion.model.bbox_hidden2output.4.weight", "diffusion.model.bbox_hidden2output.4.bias".
size mismatch for diffusion.model.init_conv.weight: copying a param with shape torch.Size([512, 512, 1]) from checkpoint, the shape in current model is torch.Size([512, 5, 1]).
size mismatch for diffusion.model.downs.0.0.mlp.1.weight: copying a param with shape torch.Size([1024, 128]) from checkpoint, the shape in current model is torch.Size([1024, 512]).
size mismatch for diffusion.model.downs.1.0.mlp.1.weight: copying a param with shape torch.Size([1024, 128]) from checkpoint, the shape in current model is torch.Size([1024, 512]).
size mismatch for diffusion.model.downs.2.0.mlp.1.weight: copying a param with shape torch.Size([1024, 128]) from checkpoint, the shape in current model is torch.Size([1024, 512]).
size mismatch for diffusion.model.downs.3.0.mlp.1.weight: copying a param with shape torch.Size([1024, 128]) from checkpoint, the shape in current model is torch.Size([1024, 512]).
size mismatch for diffusion.model.ups.0.0.mlp.1.weight: copying a param with shape torch.Size([1024, 128]) from checkpoint, the shape in current model is torch.Size([1024, 512]).
size mismatch for diffusion.model.ups.1.0.mlp.1.weight: copying a param with shape torch.Size([1024, 128]) from checkpoint, the shape in current model is torch.Size([1024, 512]).
size mismatch for diffusion.model.ups.2.0.mlp.1.weight: copying a param with shape torch.Size([1024, 128]) from checkpoint, the shape in current model is torch.Size([1024, 512]).
size mismatch for diffusion.model.ups.3.0.mlp.1.weight: copying a param with shape torch.Size([1024, 128]) from checkpoint, the shape in current model is torch.Size([1024, 512]).
size mismatch for diffusion.model.mid_block0.mlp.1.weight: copying a param with shape torch.Size([1024, 128]) from checkpoint, the shape in current model is torch.Size([1024, 512]).
This is the command I am using to run the rearrange file. "/home/ubuntu/dev/diffuscene/DiffuScene/config/rearrange/diffusion_bedrooms_instancond_lat32_v_rearrange.yaml ../pretrained/no_filtering_uncond/rearrange_top2down_notexture_nofloor /home/ubuntu/dev/diffuscene/data/sparx_internal_processed/threed_future_model_filtering.pkl --weight_file /home/ubuntu/dev/diffuscene/pretrained/bedrooms_uncond/bedrooms_uncond.pt --without_screen --n_sequences 1000 --render_top2down --save_mesh --no_texture --clip_denoised --arrange_objects"