Request checkpoints

[yhd-ai]

Request checkpoints

[SaeedSaadatnejad]

Thank you for bringing up this issue. The problem was related to training on multi-GPUs and loading on a single-GPU. Have fixed it and made updates to the code and the released model.

[SaeedSaadatnejad]

Hi,
I uploaded the trained model in the release section.
Yes, post-processing can enhance any prediction model. In the paper, we demonstrated its effectiveness with some models. Applying it to other models should be straightforward.

[yhd-ai]

Hi, I uploaded the trained model in the release section. Yes, post-processing can enhance any prediction model. In the paper, we demonstrated its effectiveness with some models. Applying it to other models should be straightforward.

I‘ve found it! However, it seems that the dict of those pertained models does not match the dict of your code. I got the error log as follows:

RuntimeError: Error(s) in loading state_dict for ModelMain:
Missing key(s) in state_dict: "embed_layer.weight", "diffmodel.diffusion_embedding.projection1.weight", "diffmodel.diffusion_embedding.projection1.bias", "diffmodel.diffusion_embedding.projection2.weight", "diffmodel.diffusion_embedding.projection2.bias", "diffmodel.input_projection.weight", "diffmodel.input_projection.bias", "diffmodel.output_projection1.weight", "diffmodel.output_projection1.bias", "diffmodel.output_projection2.weight", "diffmodel.output_projection2.bias", "diffmodel.residual_layers.0.diffusion_projection.weight", "diffmodel.residual_layers.0.diffusion_projection.bias", "diffmodel.residual_layers.0.cond_projection.weight", "diffmodel.residual_layers.0.cond_projection.bias", "diffmodel.residual_layers.0.mid_projection.weight", "diffmodel.residual_layers.0.mid_projection.bias", "diffmodel.residual_layers.0.output_projection.weight", "diffmodel.residual_layers.0.output_projection.bias", "diffmodel.residual_layers.0.time_layer.layers.0.self_attn.in_proj_weight", "diffmodel.residual_layers.0.time_layer.layers.0.self_attn.in_proj_bias", "diffmodel.residual_layers.0.time_layer.layers.0.self_attn.out_proj.weight", "diffmodel.residual_layers.0.time_layer.layers.0.self_attn.out_proj.bias", "diffmodel.residual_layers.0.time_layer.layers.0.linear1.weight", "diffmodel.residual_layers.0.time_layer.layers.0.linear1.bias", "diffmodel.residual_layers.0.time_layer.layers.0.linear2.weight", "diffmodel.residual_layers.0.time_layer.layers.0.linear2.bias", "diffmodel.residual_layers.0.time_layer.layers.0.norm1.weight", "diffmodel.residual_layers.0.time_layer.layers.0.norm1.bias", "diffmodel.residual_layers.0.time_layer.layers.0.norm2.weight", "diffmodel.residual_layers.0.time_layer.layers.0.norm2.bias", "diffmodel.residual_layers.0.feature_layer.layers.0.self_attn.in_proj_weight", "diffmodel.residual_layers.0.feature_layer.layers.0.self_attn.in_proj_bias", "diffmodel.residual_layers.0.feature_layer.layers.0.self_attn.out_proj.weight", "diffmodel.residual_layers.0.feature_layer.layers.0.self_attn.out_proj.bias", "diffmodel.residual_layers.0.feature_layer.layers.0.linear1.weight", "diffmodel.residual_layers.0.feature_layer.layers.0.linear1.bias", "diffmodel.residual_layers.0.feature_layer.layers.0.linear2.weight", "diffmodel.residual_layers.0.feature_layer.layers.0.linear2.bias", "diffmodel.residual_layers.0.feature_layer.layers.0.norm1.weight", "diffmodel.residual_layers.0.feature_layer.layers.0.norm1.bias", "diffmodel.residual_layers.0.feature_layer.layers.0.norm2.weight", "diffmodel.residual_layers.0.feature_layer.layers.0.norm2.bias", "diffmodel.residual_layers.1.diffusion_projection.weight", "diffmodel.residual_layers.1.diffusion_projection.bias", "diffmodel.residual_layers.1.cond_projection.weight", "diffmodel.residual_layers.1.cond_projection.bias", "diffmodel.residual_layers.1.mid_projection.weight", "diffmodel.residual_layers.1.mid_projection.bias", "diffmodel.residual_layers.1.output_projection.weight", "diffmodel.residual_layers.1.output_projection.bias", "diffmodel.residual_layers.1.time_layer.layers.0.self_attn.in_proj_weight", "diffmodel.residual_layers.1.time_layer.layers.0.self_attn.in_proj_bias", "diffmodel.residual_layers.1.time_layer.layers.0.self_attn.out_proj.weight", "diffmodel.residual_layers.1.time_layer.layers.0.self_attn.out_proj.bias", "diffmodel.residual_layers.1.time_layer.layers.0.linear1.weight", "diffmodel.residual_layers.1.time_layer.layers.0.linear1.bias", "diffmodel.residual_layers.1.time_layer.layers.0.linear2.weight", "diffmodel.residual_layers.1.time_layer.layers.0.linear2.bias", "diffmodel.residual_layers.1.time_layer.layers.0.norm1.weight", "diffmodel.residual_layers.1.time_layer.layers.0.norm1.bias", "diffmodel.residual_layers.1.time_layer.layers.0.norm2.weight", "diffmodel.residual_layers.1.time_layer.layers.0.norm2.bias", "diffmodel.residual_layers.1.feature_layer.layers.0.self_attn.in_proj_weight", "diffmodel.residual_layers.1.feature_layer.layers.0.self_attn.in_proj_bias", "diffmodel.residual_layers.1.feature_layer.layers.0.self_attn.out_proj.weight", "diffmodel.residual_layers.1.feature_layer.layers.0.self_attn.out_proj.bias", "diffmodel.residual_layers.1.feature_layer.layers.0.linear1.weight", "diffmodel.residual_layers.1.feature_layer.layers.0.linear1.bias", "diffmodel.residual_layers.1.feature_layer.layers.0.linear2.weight", "diffmodel.residual_layers.1.feature_layer.layers.0.linear2.bias", "diffmodel.residual_layers.1.feature_layer.layers.0.norm1.weight", "diffmodel.residual_layers.1.feature_layer.layers.0.norm1.bias", "diffmodel.residual_layers.1.feature_layer.layers.0.norm2.weight", "diffmodel.residual_layers.1.feature_layer.layers.0.norm2.bias", "diffmodel.residual_layers.2.diffusion_projection.weight", "diffmodel.residual_layers.2.diffusion_projection.bias", "diffmodel.residual_layers.2.cond_projection.weight", "diffmodel.residual_layers.2.cond_projection.bias", "diffmodel.residual_layers.2.mid_projection.weight", "diffmodel.residual_layers.2.mid_projection.bias", "diffmodel.residual_layers.2.output_projection.weight", "diffmodel.residual_layers.2.output_projection.bias", "diffmodel.residual_layers.2.time_layer.layers.0.self_attn.in_proj_weight", "diffmodel.residual_layers.2.time_layer.layers.0.self_attn.in_proj_bias", "diffmodel.residual_layers.2.time_layer.layers.0.self_attn.out_proj.weight", "diffmodel.residual_layers.2.time_layer.layers.0.self_attn.out_proj.bias", "diffmodel.residual_layers.2.time_layer.layers.0.linear1.weight", "diffmodel.residual_layers.2.time_layer.layers.0.linear1.bias", "diffmodel.residual_layers.2.time_layer.layers.0.linear2.weight", "diffmodel.residual_layers.2.time_layer.layers.0.linear2.bias", "diffmodel.residual_layers.2.time_layer.layers.0.norm1.weight", "diffmodel.residual_layers.2.time_layer.layers.0.norm1.bias", "diffmodel.residual_layers.2.time_layer.layers.0.norm2.weight", "diffmodel.residual_layers.2.time_layer.layers.0.norm2.bias", "diffmodel.residual_layers.2.feature_layer.layers.0.self_attn.in_proj_weight", "diffmodel.residual_layers.2.feature_layer.layers.0.self_attn.in_proj_bias", "diffmodel.residual_layers.2.feature_layer.layers.0.self_attn.out_proj.weight", "diffmodel.residual_layers.2.feature_layer.layers.0.self_attn.out_proj.bias", "diffmodel.residual_layers.2.feature_layer.layers.0.linear1.weight", "diffmodel.residual_layers.2.feature_layer.layers.0.linear1.bias", "diffmodel.residual_layers.2.feature_layer.layers.0.linear2.weight", "diffmodel.residual_layers.2.feature_layer.layers.0.linear2.bias", "diffmodel.residual_layers.2.feature_layer.layers.0.norm1.weight", "diffmodel.residual_layers.2.feature_layer.layers.0.norm1.bias", "diffmodel.residual_layers.2.feature_layer.layers.0.norm2.weight", "diffmodel.residual_layers.2.feature_layer.layers.0.norm2.bias", "diffmodel.residual_layers.3.diffusion_projection.weight", "diffmodel.residual_layers.3.diffusion_projection.bias", "diffmodel.residual_layers.3.cond_projection.weight", "diffmodel.residual_layers.3.cond_projection.bias", "diffmodel.residual_layers.3.mid_projection.weight", "diffmodel.residual_layers.3.mid_projection.bias", "diffmodel.residual_layers.3.output_projection.weight", "diffmodel.residual_layers.3.output_projection.bias", "diffmodel.residual_layers.3.time_layer.layers.0.self_attn.in_proj_weight", "diffmodel.residual_layers.3.time_layer.layers.0.self_attn.in_proj_bias", "diffmodel.residual_layers.3.time_layer.layers.0.self_attn.out_proj.weight", "diffmodel.residual_layers.3.time_layer.layers.0.self_attn.out_proj.bias", "diffmodel.residual_layers.3.time_layer.layers.0.linear1.weight", "diffmodel.residual_layers.3.time_layer.layers.0.linear1.bias", "diffmodel.residual_layers.3.time_layer.layers.0.linear2.weight", "diffmodel.residual_layers.3.time_layer.layers.0.linear2.bias", "diffmodel.residual_layers.3.time_layer.layers.0.norm1.weight", "diffmodel.residual_layers.3.time_layer.layers.0.norm1.bias", "diffmodel.residual_layers.3.time_layer.layers.0.norm2.weight", "diffmodel.residual_layers.3.time_layer.layers.0.norm2.bias", "diffmodel.residual_layers.3.feature_layer.layers.0.self_attn.in_proj_weight", "diffmodel.residual_layers.3.feature_layer.layers.0.self_attn.in_proj_bias", "diffmodel.residual_layers.3.feature_layer.layers.0.self_attn.out_proj.weight", "diffmodel.residual_layers.3.feature_layer.layers.0.self_attn.out_proj.bias", "diffmodel.residual_layers.3.feature_layer.layers.0.linear1.weight", "diffmodel.residual_layers.3.feature_layer.layers.0.linear1.bias", "diffmodel.residual_layers.3.feature_layer.layers.0.linear2.weight", "diffmodel.residual_layers.3.feature_layer.layers.0.linear2.bias", "diffmodel.residual_layers.3.feature_layer.layers.0.norm1.weight", "diffmodel.residual_layers.3.feature_layer.layers.0.norm1.bias", "diffmodel.residual_layers.3.feature_layer.layers.0.norm2.weight", "diffmodel.residual_layers.3.feature_layer.layers.0.norm2.bias", "diffmodel.residual_layers.4.diffusion_projection.weight", "diffmodel.residual_layers.4.diffusion_projection.bias", "diffmodel.residual_layers.4.cond_projection.weight", "diffmodel.residual_layers.4.cond_projection.bias", "diffmodel.residual_layers.4.mid_projection.weight", "diffmodel.residual_layers.4.mid_projection.bias", "diffmodel.residual_layers.4.output_projection.weight", "diffmodel.residual_layers.4.output_projection.bias", "diffmodel.residual_layers.4.time_layer.layers.0.self_attn.in_proj_weight", "diffmodel.residual_layers.4.time_layer.layers.0.self_attn.in_proj_bias", "diffmodel.residual_layers.4.time_layer.layers.0.self_attn.out_proj.weight", "diffmodel.residual_layers.4.time_layer.layers.0.self_attn.out_proj.bias", "diffmodel.residual_layers.4.time_layer.layers.0.linear1.weight", "diffmodel.residual_layers.4.time_layer.layers.0.linear1.bias", "diffmodel.residual_layers.4.time_layer.layers.0.linear2.weight", "diffmodel.residual_layers.4.time_layer.layers.0.linear2.bias", "diffmodel.residual_layers.4.time_layer.layers.0.norm1.weight", "diffmodel.residual_layers.4.time_layer.layers.0.norm1.bias", "diffmodel.residual_layers.4.time_layer.layers.0.norm2.weight", "diffmodel.residual_layers.4.time_layer.layers.0.norm2.bias", "diffmodel.residual_layers.4.feature_layer.layers.0.self_attn.in_proj_weight", "diffmodel.residual_layers.4.feature_layer.layers.0.self_attn.in_proj_bias", "diffmodel.residual_layers.4.feature_layer.layers.0.self_attn.out_proj.weight", "diffmodel.residual_layers.4.feature_layer.layers.0.self_attn.out_proj.bias", "diffmodel.residual_layers.4.feature_layer.layers.0.linear1.weight", "diffmodel.residual_layers.4.feature_layer.layers.0.linear1.bias", "diffmodel.residual_layers.4.feature_layer.layers.0.linear2.weight", "diffmodel.residual_layers.4.feature_layer.layers.0.linear2.bias", "diffmodel.residual_layers.4.feature_layer.layers.0.norm1.weight", "diffmodel.residual_layers.4.feature_layer.layers.0.norm1.bias", "diffmodel.residual_layers.4.feature_layer.layers.0.norm2.weight", "diffmodel.residual_layers.4.feature_layer.layers.0.norm2.bias", "diffmodel.residual_layers.5.diffusion_projection.weight", "diffmodel.residual_layers.5.diffusion_projection.bias", "diffmodel.residual_layers.5.cond_projection.weight", "diffmodel.residual_layers.5.cond_projection.bias", "diffmodel.residual_layers.5.mid_projection.weight", "diffmodel.residual_layers.5.mid_projection.bias", "diffmodel.residual_layers.5.output_projection.weight", "diffmodel.residual_layers.5.output_projection.bias", "diffmodel.residual_layers.5.time_layer.layers.0.self_attn.in_proj_weight", "diffmodel.residual_layers.5.time_layer.layers.0.self_attn.in_proj_bias", "diffmodel.residual_layers.5.time_layer.layers.0.self_attn.out_proj.weight", "diffmodel.residual_layers.5.time_layer.layers.0.self_attn.out_proj.bias", "diffmodel.residual_layers.5.time_layer.layers.0.linear1.weight", "diffmodel.residual_layers.5.time_layer.layers.0.linear1.bias", "diffmodel.residual_layers.5.time_layer.layers.0.linear2.weight", "diffmodel.residual_layers.5.time_layer.layers.0.linear2.bias", "diffmodel.residual_layers.5.time_layer.layers.0.norm1.weight", "diffmodel.residual_layers.5.time_layer.layers.0.norm1.bias", "diffmodel.residual_layers.5.time_layer.layers.0.norm2.weight", "diffmodel.residual_layers.5.time_layer.layers.0.norm2.bias", "diffmodel.residual_layers.5.feature_layer.layers.0.self_attn.in_proj_weight", "diffmodel.residual_layers.5.feature_layer.layers.0.self_attn.in_proj_bias", "diffmodel.residual_layers.5.feature_layer.layers.0.self_attn.out_proj.weight", "diffmodel.residual_layers.5.feature_layer.layers.0.self_attn.out_proj.bias", "diffmodel.residual_layers.5.feature_layer.layers.0.linear1.weight", "diffmodel.residual_layers.5.feature_layer.layers.0.linear1.bias", "diffmodel.residual_layers.5.feature_layer.layers.0.linear2.weight", "diffmodel.residual_layers.5.feature_layer.layers.0.linear2.bias", "diffmodel.residual_layers.5.feature_layer.layers.0.norm1.weight", "diffmodel.residual_layers.5.feature_layer.layers.0.norm1.bias", "diffmodel.residual_layers.5.feature_layer.layers.0.norm2.weight", "diffmodel.residual_layers.5.feature_layer.layers.0.norm2.bias", "diffmodel.residual_layers.6.diffusion_projection.weight", "diffmodel.residual_layers.6.diffusion_projection.bias", "diffmodel.residual_layers.6.cond_projection.weight", "diffmodel.residual_layers.6.cond_projection.bias", "diffmodel.residual_layers.6.mid_projection.weight", "diffmodel.residual_layers.6.mid_projection.bias", "diffmodel.residual_layers.6.output_projection.weight", "diffmodel.residual_layers.6.output_projection.bias", "diffmodel.residual_layers.6.time_layer.layers.0.self_attn.in_proj_weight", "diffmodel.residual_layers.6.time_layer.layers.0.self_attn.in_proj_bias", "diffmodel.residual_layers.6.time_layer.layers.0.self_attn.out_proj.weight", "diffmodel.residual_layers.6.time_layer.layers.0.self_attn.out_proj.bias", "diffmodel.residual_layers.6.time_layer.layers.0.linear1.weight", "diffmodel.residual_layers.6.time_layer.layers.0.linear1.bias", "diffmodel.residual_layers.6.time_layer.layers.0.linear2.weight", "diffmodel.residual_layers.6.time_layer.layers.0.linear2.bias", "diffmodel.residual_layers.6.time_layer.layers.0.norm1.weight", "diffmodel.residual_layers.6.time_layer.layers.0.norm1.bias", "diffmodel.residual_layers.6.time_layer.layers.0.norm2.weight", "diffmodel.residual_layers.6.time_layer.layers.0.norm2.bias", "diffmodel.residual_layers.6.feature_layer.layers.0.self_attn.in_proj_weight", "diffmodel.residual_layers.6.feature_layer.layers.0.self_attn.in_proj_bias", "diffmodel.residual_layers.6.feature_layer.layers.0.self_attn.out_proj.weight", "diffmodel.residual_layers.6.feature_layer.layers.0.self_attn.out_proj.bias", "diffmodel.residual_layers.6.feature_layer.layers.0.linear1.weight", "diffmodel.residual_layers.6.feature_layer.layers.0.linear1.bias", "diffmodel.residual_layers.6.feature_layer.layers.0.linear2.weight", "diffmodel.residual_layers.6.feature_layer.layers.0.linear2.bias", "diffmodel.residual_layers.6.feature_layer.layers.0.norm1.weight", "diffmodel.residual_layers.6.feature_layer.layers.0.norm1.bias", "diffmodel.residual_layers.6.feature_layer.layers.0.norm2.weight", "diffmodel.residual_layers.6.feature_layer.layers.0.norm2.bias", "diffmodel.residual_layers.7.diffusion_projection.weight", "diffmodel.residual_layers.7.diffusion_projection.bias", "diffmodel.residual_layers.7.cond_projection.weight", "diffmodel.residual_layers.7.cond_projection.bias", "diffmodel.residual_layers.7.mid_projection.weight", "diffmodel.residual_layers.7.mid_projection.bias", "diffmodel.residual_layers.7.output_projection.weight", "diffmodel.residual_layers.7.output_projection.bias", "diffmodel.residual_layers.7.time_layer.layers.0.self_attn.in_proj_weight", "diffmodel.residual_layers.7.time_layer.layers.0.self_attn.in_proj_bias", "diffmodel.residual_layers.7.time_layer.layers.0.self_attn.out_proj.weight", "diffmodel.residual_layers.7.time_layer.layers.0.self_attn.out_proj.bias", "diffmodel.residual_layers.7.time_layer.layers.0.linear1.weight", "diffmodel.residual_layers.7.time_layer.layers.0.linear1.bias", "diffmodel.residual_layers.7.time_layer.layers.0.linear2.weight", "diffmodel.residual_layers.7.time_layer.layers.0.linear2.bias", "diffmodel.residual_layers.7.time_layer.layers.0.norm1.weight", "diffmodel.residual_layers.7.time_layer.layers.0.norm1.bias", "diffmodel.residual_layers.7.time_layer.layers.0.norm2.weight", "diffmodel.residual_layers.7.time_layer.layers.0.norm2.bias", "diffmodel.residual_layers.7.feature_layer.layers.0.self_attn.in_proj_weight", "diffmodel.residual_layers.7.feature_layer.layers.0.self_attn.in_proj_bias", "diffmodel.residual_layers.7.feature_layer.layers.0.self_attn.out_proj.weight", "diffmodel.residual_layers.7.feature_layer.layers.0.self_attn.out_proj.bias", "diffmodel.residual_layers.7.feature_layer.layers.0.linear1.weight", "diffmodel.residual_layers.7.feature_layer.layers.0.linear1.bias", "diffmodel.residual_layers.7.feature_layer.layers.0.linear2.weight", "diffmodel.residual_layers.7.feature_layer.layers.0.linear2.bias", "diffmodel.residual_layers.7.feature_layer.layers.0.norm1.weight", "diffmodel.residual_layers.7.feature_layer.layers.0.norm1.bias", "diffmodel.residual_layers.7.feature_layer.layers.0.norm2.weight", "diffmodel.residual_layers.7.feature_layer.layers.0.norm2.bias", "diffmodel.residual_layers.8.diffusion_projection.weight", "diffmodel.residual_layers.8.diffusion_projection.bias", "diffmodel.residual_layers.8.cond_projection.weight", "diffmodel.residual_layers.8.cond_projection.bias", "diffmodel.residual_layers.8.mid_projection.weight", "diffmodel.residual_layers.8.mid_projection.bias", "diffmodel.residual_layers.8.output_projection.weight", "diffmodel.residual_layers.8.output_projection.bias", "diffmodel.residual_layers.8.time_layer.layers.0.self_attn.in_proj_weight", "diffmodel.residual_layers.8.time_layer.layers.0.self_attn.in_proj_bias", "diffmodel.residual_layers.8.time_layer.layers.0.self_attn.out_proj.weight", "diffmodel.residual_layers.8.time_layer.layers.0.self_attn.out_proj.bias", "diffmodel.residual_layers.8.time_layer.layers.0.linear1.weight", "diffmodel.residual_layers.8.time_layer.layers.0.linear1.bias", "diffmodel.residual_layers.8.time_layer.layers.0.linear2.weight", "diffmodel.residual_layers.8.time_layer.layers.0.linear2.bias", "diffmodel.residual_layers.8.time_layer.layers.0.norm1.weight", "diffmodel.residual_layers.8.time_layer.layers.0.norm1.bias", "diffmodel.residual_layers.8.time_layer.layers.0.norm2.weight", "diffmodel.residual_layers.8.time_layer.layers.0.norm2.bias", "diffmodel.residual_layers.8.feature_layer.layers.0.self_attn.in_proj_weight", "diffmodel.residual_layers.8.feature_layer.layers.0.self_attn.in_proj_bias", "diffmodel.residual_layers.8.feature_layer.layers.0.self_attn.out_proj.weight", "diffmodel.residual_layers.8.feature_layer.layers.0.self_attn.out_proj.bias", "diffmodel.residual_layers.8.feature_layer.layers.0.linear1.weight", "diffmodel.residual_layers.8.feature_layer.layers.0.linear1.bias", "diffmodel.residual_layers.8.feature_layer.layers.0.linear2.weight", "diffmodel.residual_layers.8.feature_layer.layers.0.linear2.bias", "diffmodel.residual_layers.8.feature_layer.layers.0.norm1.weight", "diffmodel.residual_layers.8.feature_layer.layers.0.norm1.bias", "diffmodel.residual_layers.8.feature_layer.layers.0.norm2.weight", "diffmodel.residual_layers.8.feature_layer.layers.0.norm2.bias", "diffmodel.residual_layers.9.diffusion_projection.weight", "diffmodel.residual_layers.9.diffusion_projection.bias", "diffmodel.residual_layers.9.cond_projection.weight", "diffmodel.residual_layers.9.cond_projection.bias", "diffmodel.residual_layers.9.mid_projection.weight", "diffmodel.residual_layers.9.mid_projection.bias", "diffmodel.residual_layers.9.output_projection.weight", "diffmodel.residual_layers.9.output_projection.bias", "diffmodel.residual_layers.9.time_layer.layers.0.self_attn.in_proj_weight", "diffmodel.residual_layers.9.time_layer.layers.0.self_attn.in_proj_bias", "diffmodel.residual_layers.9.time_layer.layers.0.self_attn.out_proj.weight", "diffmodel.residual_layers.9.time_layer.layers.0.self_attn.out_proj.bias", "diffmodel.residual_layers.9.time_layer.layers.0.linear1.weight", "diffmodel.residual_layers.9.time_layer.layers.0.linear1.bias", "diffmodel.residual_layers.9.time_layer.layers.0.linear2.weight", "diffmodel.residual_layers.9.time_layer.layers.0.linear2.bias", "diffmodel.residual_layers.9.time_layer.layers.0.norm1.weight", "diffmodel.residual_layers.9.time_layer.layers.0.norm1.bias", "diffmodel.residual_layers.9.time_layer.layers.0.norm2.weight", "diffmodel.residual_layers.9.time_layer.layers.0.norm2.bias", "diffmodel.residual_layers.9.feature_layer.layers.0.self_attn.in_proj_weight", "diffmodel.residual_layers.9.feature_layer.layers.0.self_attn.in_proj_bias", "diffmodel.residual_layers.9.feature_layer.layers.0.self_attn.out_proj.weight", "diffmodel.residual_layers.9.feature_layer.layers.0.self_attn.out_proj.bias", "diffmodel.residual_layers.9.feature_layer.layers.0.linear1.weight", "diffmodel.residual_layers.9.feature_layer.layers.0.linear1.bias", "diffmodel.residual_layers.9.feature_layer.layers.0.linear2.weight", "diffmodel.residual_layers.9.feature_layer.layers.0.linear2.bias", "diffmodel.residual_layers.9.feature_layer.layers.0.norm1.weight", "diffmodel.residual_layers.9.feature_layer.layers.0.norm1.bias", "diffmodel.residual_layers.9.feature_layer.layers.0.norm2.weight", "diffmodel.residual_layers.9.feature_layer.layers.0.norm2.bias", "diffmodel.residual_layers.10.diffusion_projection.weight", "diffmodel.residual_layers.10.diffusion_projection.bias", "diffmodel.residual_layers.10.cond_projection.weight", "diffmodel.residual_layers.10.cond_projection.bias", "diffmodel.residual_layers.10.mid_projection.weight", "diffmodel.residual_layers.10.mid_projection.bias", "diffmodel.residual_layers.10.output_projection.weight", "diffmodel.residual_layers.10.output_projection.bias", "diffmodel.residual_layers.10.time_layer.layers.0.self_attn.in_proj_weight", "diffmodel.residual_layers.10.time_layer.layers.0.self_attn.in_proj_bias", "diffmodel.residual_layers.10.time_layer.layers.0.self_attn.out_proj.weight", "diffmodel.residual_layers.10.time_layer.layers.0.self_attn.out_proj.bias", "diffmodel.residual_layers.10.time_layer.layers.0.linear1.weight", "diffmodel.residual_layers.10.time_layer.layers.0.linear1.bias", "diffmodel.residual_layers.10.time_layer.layers.0.linear2.weight", "diffmodel.residual_layers.10.time_layer.layers.0.linear2.bias", "diffmodel.residual_layers.10.time_layer.layers.0.norm1.weight", "diffmodel.residual_layers.10.time_layer.layers.0.norm1.bias", "diffmodel.residual_layers.10.time_layer.layers.0.norm2.weight", "diffmodel.residual_layers.10.time_layer.layers.0.norm2.bias", "diffmodel.residual_layers.10.feature_layer.layers.0.self_attn.in_proj_weight", "diffmodel.residual_layers.10.feature_layer.layers.0.self_attn.in_proj_bias", "diffmodel.residual_layers.10.feature_layer.layers.0.self_attn.out_proj.weight", "diffmodel.residual_layers.10.feature_layer.layers.0.self_attn.out_proj.bias", "diffmodel.residual_layers.10.feature_layer.layers.0.linear1.weight", "diffmodel.residual_layers.10.feature_layer.layers.0.linear1.bias", "diffmodel.residual_layers.10.feature_layer.layers.0.linear2.weight", "diffmodel.residual_layers.10.feature_layer.layers.0.linear2.bias", "diffmodel.residual_layers.10.feature_layer.layers.0.norm1.weight", "diffmodel.residual_layers.10.feature_layer.layers.0.norm1.bias", "diffmodel.residual_layers.10.feature_layer.layers.0.norm2.weight", "diffmodel.residual_layers.10.feature_layer.layers.0.norm2.bias", "diffmodel.residual_layers.11.diffusion_projection.weight", "diffmodel.residual_layers.11.diffusion_projection.bias", "diffmodel.residual_layers.11.cond_projection.weight", "diffmodel.residual_layers.11.cond_projection.bias", "diffmodel.residual_layers.11.mid_projection.weight", "diffmodel.residual_layers.11.mid_projection.bias", "diffmodel.residual_layers.11.output_projection.weight", "diffmodel.residual_layers.11.output_projection.bias", "diffmodel.residual_layers.11.time_layer.layers.0.self_attn.in_proj_weight", "diffmodel.residual_layers.11.time_layer.layers.0.self_attn.in_proj_bias", "diffmodel.residual_layers.11.time_layer.layers.0.self_attn.out_proj.weight", "diffmodel.residual_layers.11.time_layer.layers.0.self_attn.out_proj.bias", "diffmodel.residual_layers.11.time_layer.layers.0.linear1.weight", "diffmodel.residual_layers.11.time_layer.layers.0.linear1.bias", "diffmodel.residual_layers.11.time_layer.layers.0.linear2.weight", "diffmodel.residual_layers.11.time_layer.layers.0.linear2.bias", "diffmodel.residual_layers.11.time_layer.layers.0.norm1.weight", "diffmodel.residual_layers.11.time_layer.layers.0.norm1.bias", "diffmodel.residual_layers.11.time_layer.layers.0.norm2.weight", "diffmodel.residual_layers.11.time_layer.layers.0.norm2.bias", "diffmodel.residual_layers.11.feature_layer.layers.0.self_attn.in_proj_weight", "diffmodel.residual_layers.11.feature_layer.layers.0.self_attn.in_proj_bias", "diffmodel.residual_layers.11.feature_layer.layers.0.self_attn.out_proj.weight", "diffmodel.residual_layers.11.feature_layer.layers.0.self_attn.out_proj.bias", "diffmodel.residual_layers.11.feature_layer.layers.0.linear1.weight", "diffmodel.residual_layers.11.feature_layer.layers.0.linear1.bias", "diffmodel.residual_layers.11.feature_layer.layers.0.linear2.weight", "diffmodel.residual_layers.11.feature_layer.layers.0.linear2.bias", "diffmodel.residual_layers.11.feature_layer.layers.0.norm1.weight", "diffmodel.residual_layers.11.feature_layer.layers.0.norm1.bias", "diffmodel.residual_layers.11.feature_layer.layers.0.norm2.weight", "diffmodel.residual_layers.11.feature_layer.layers.0.norm2.bias".
Unexpected key(s) in state_dict: "module.embed_layer.weight", "module.diffmodel.diffusion_embedding.projection1.weight", "module.diffmodel.diffusion_embedding.projection1.bias", "module.diffmodel.diffusion_embedding.projection2.weight", "module.diffmodel.diffusion_embedding.projection2.bias", "module.diffmodel.input_projection.weight", "module.diffmodel.input_projection.bias", "module.diffmodel.output_projection1.weight", "module.diffmodel.output_projection1.bias", "module.diffmodel.output_projection2.weight", "module.diffmodel.output_projection2.bias", "module.diffmodel.residual_layers.0.diffusion_projection.weight", "module.diffmodel.residual_layers.0.diffusion_projection.bias", "module.diffmodel.residual_layers.0.cond_projection.weight", "module.diffmodel.residual_layers.0.cond_projection.bias", "module.diffmodel.residual_layers.0.mid_projection.weight", "module.diffmodel.residual_layers.0.mid_projection.bias", "module.diffmodel.residual_layers.0.output_projection.weight", "module.diffmodel.residual_layers.0.output_projection.bias", "module.diffmodel.residual_layers.0.time_layer.layers.0.self_attn.in_proj_weight", "module.diffmodel.residual_layers.0.time_layer.layers.0.self_attn.in_proj_bias", "module.diffmodel.residual_layers.0.time_layer.layers.0.self_attn.out_proj.weight", "module.diffmodel.residual_layers.0.time_layer.layers.0.self_attn.out_proj.bias", "module.diffmodel.residual_layers.0.time_layer.layers.0.linear1.weight", "module.diffmodel.residual_layers.0.time_layer.layers.0.linear1.bias", "module.diffmodel.residual_layers.0.time_layer.layers.0.linear2.weight", "module.diffmodel.residual_layers.0.time_layer.layers.0.linear2.bias", "module.diffmodel.residual_layers.0.time_layer.layers.0.norm1.weight", "module.diffmodel.residual_layers.0.time_layer.layers.0.norm1.bias", "module.diffmodel.residual_layers.0.time_layer.layers.0.norm2.weight", "module.diffmodel.residual_layers.0.time_layer.layers.0.norm2.bias", "module.diffmodel.residual_layers.0.feature_layer.layers.0.self_attn.in_proj_weight", "module.diffmodel.residual_layers.0.feature_layer.layers.0.self_attn.in_proj_bias", "module.diffmodel.residual_layers.0.feature_layer.layers.0.self_attn.out_proj.weight", "module.diffmodel.residual_layers.0.feature_layer.layers.0.self_attn.out_proj.bias", "module.diffmodel.residual_layers.0.feature_layer.layers.0.linear1.weight", "module.diffmodel.residual_layers.0.feature_layer.layers.0.linear1.bias", "module.diffmodel.residual_layers.0.feature_layer.layers.0.linear2.weight", "module.diffmodel.residual_layers.0.feature_layer.layers.0.linear2.bias", "module.diffmodel.residual_layers.0.feature_layer.layers.0.norm1.weight", "module.diffmodel.residual_layers.0.feature_layer.layers.0.norm1.bias", "module.diffmodel.residual_layers.0.feature_layer.layers.0.norm2.weight", "module.diffmodel.residual_layers.0.feature_layer.layers.0.norm2.bias", "module.diffmodel.residual_layers.1.diffusion_projection.weight", "module.diffmodel.residual_layers.1.diffusion_projection.bias", "module.diffmodel.residual_layers.1.cond_projection.weight", "module.diffmodel.residual_layers.1.cond_projection.bias", "module.diffmodel.residual_layers.1.mid_projection.weight", "module.diffmodel.residual_layers.1.mid_projection.bias", "module.diffmodel.residual_layers.1.output_projection.weight", "module.diffmodel.residual_layers.1.output_projection.bias", "module.diffmodel.residual_layers.1.time_layer.layers.0.self_attn.in_proj_weight", "module.diffmodel.residual_layers.1.time_layer.layers.0.self_attn.in_proj_bias", "module.diffmodel.residual_layers.1.time_layer.layers.0.self_attn.out_proj.weight", "module.diffmodel.residual_layers.1.time_layer.layers.0.self_attn.out_proj.bias", "module.diffmodel.residual_layers.1.time_layer.layers.0.linear1.weight", "module.diffmodel.residual_layers.1.time_layer.layers.0.linear1.bias", "module.diffmodel.residual_layers.1.time_layer.layers.0.linear2.weight", "module.diffmodel.residual_layers.1.time_layer.layers.0.linear2.bias", "module.diffmodel.residual_layers.1.time_layer.layers.0.norm1.weight", "module.diffmodel.residual_layers.1.time_layer.layers.0.norm1.bias", "module.diffmodel.residual_layers.1.time_layer.layers.0.norm2.weight", "module.diffmodel.residual_layers.1.time_layer.layers.0.norm2.bias", "module.diffmodel.residual_layers.1.feature_layer.layers.0.self_attn.in_proj_weight", "module.diffmodel.residual_layers.1.feature_layer.layers.0.self_attn.in_proj_bias", "module.diffmodel.residual_layers.1.feature_layer.layers.0.self_attn.out_proj.weight", "module.diffmodel.residual_layers.1.feature_layer.layers.0.self_attn.out_proj.bias", "module.diffmodel.residual_layers.1.feature_layer.layers.0.linear1.weight", "module.diffmodel.residual_layers.1.feature_layer.layers.0.linear1.bias", "module.diffmodel.residual_layers.1.feature_layer.layers.0.linear2.weight", "module.diffmodel.residual_layers.1.feature_layer.layers.0.linear2.bias", "module.diffmodel.residual_layers.1.feature_layer.layers.0.norm1.weight", "module.diffmodel.residual_layers.1.feature_layer.layers.0.norm1.bias", "module.diffmodel.residual_layers.1.feature_layer.layers.0.norm2.weight", "module.diffmodel.residual_layers.1.feature_layer.layers.0.norm2.bias", "module.diffmodel.residual_layers.2.diffusion_projection.weight", "module.diffmodel.residual_layers.2.diffusion_projection.bias", "module.diffmodel.residual_layers.2.cond_projection.weight", "module.diffmodel.residual_layers.2.cond_projection.bias", "module.diffmodel.residual_layers.2.mid_projection.weight", "module.diffmodel.residual_layers.2.mid_projection.bias", "module.diffmodel.residual_layers.2.output_projection.weight", "module.diffmodel.residual_layers.2.output_projection.bias", "module.diffmodel.residual_layers.2.time_layer.layers.0.self_attn.in_proj_weight", "module.diffmodel.residual_layers.2.time_layer.layers.0.self_attn.in_proj_bias", "module.diffmodel.residual_layers.2.time_layer.layers.0.self_attn.out_proj.weight", "module.diffmodel.residual_layers.2.time_layer.layers.0.self_attn.out_proj.bias", "module.diffmodel.residual_layers.2.time_layer.layers.0.linear1.weight", "module.diffmodel.residual_layers.2.time_layer.layers.0.linear1.bias", "module.diffmodel.residual_layers.2.time_layer.layers.0.linear2.weight", "module.diffmodel.residual_layers.2.time_layer.layers.0.linear2.bias", "module.diffmodel.residual_layers.2.time_layer.layers.0.norm1.weight", "module.diffmodel.residual_layers.2.time_layer.layers.0.norm1.bias", "module.diffmodel.residual_layers.2.time_layer.layers.0.norm2.weight", "module.diffmodel.residual_layers.2.time_layer.layers.0.norm2.bias", "module.diffmodel.residual_layers.2.feature_layer.layers.0.self_attn.in_proj_weight", "module.diffmodel.residual_layers.2.feature_layer.layers.0.self_attn.in_proj_bias", "module.diffmodel.residual_layers.2.feature_layer.layers.0.self_attn.out_proj.weight", "module.diffmodel.residual_layers.2.feature_layer.layers.0.self_attn.out_proj.bias", "module.diffmodel.residual_layers.2.feature_layer.layers.0.linear1.weight", "module.diffmodel.residual_layers.2.feature_layer.layers.0.linear1.bias", "module.diffmodel.residual_layers.2.feature_layer.layers.0.linear2.weight", "module.diffmodel.residual_layers.2.feature_layer.layers.0.linear2.bias", "module.diffmodel.residual_layers.2.feature_layer.layers.0.norm1.weight", "module.diffmodel.residual_layers.2.feature_layer.layers.0.norm1.bias", "module.diffmodel.residual_layers.2.feature_layer.layers.0.norm2.weight", "module.diffmodel.residual_layers.2.feature_layer.layers.0.norm2.bias", "module.diffmodel.residual_layers.3.diffusion_projection.weight", "module.diffmodel.residual_layers.3.diffusion_projection.bias", "module.diffmodel.residual_layers.3.cond_projection.weight", "module.diffmodel.residual_layers.3.cond_projection.bias", "module.diffmodel.residual_layers.3.mid_projection.weight", "module.diffmodel.residual_layers.3.mid_projection.bias", "module.diffmodel.residual_layers.3.output_projection.weight", "module.diffmodel.residual_layers.3.output_projection.bias", "module.diffmodel.residual_layers.3.time_layer.layers.0.self_attn.in_proj_weight", "module.diffmodel.residual_layers.3.time_layer.layers.0.self_attn.in_proj_bias", "module.diffmodel.residual_layers.3.time_layer.layers.0.self_attn.out_proj.weight", "module.diffmodel.residual_layers.3.time_layer.layers.0.self_attn.out_proj.bias", "module.diffmodel.residual_layers.3.time_layer.layers.0.linear1.weight", "module.diffmodel.residual_layers.3.time_layer.layers.0.linear1.bias", "module.diffmodel.residual_layers.3.time_layer.layers.0.linear2.weight", "module.diffmodel.residual_layers.3.time_layer.layers.0.linear2.bias", "module.diffmodel.residual_layers.3.time_layer.layers.0.norm1.weight", "module.diffmodel.residual_layers.3.time_layer.layers.0.norm1.bias", "module.diffmodel.residual_layers.3.time_layer.layers.0.norm2.weight", "module.diffmodel.residual_layers.3.time_layer.layers.0.norm2.bias", "module.diffmodel.residual_layers.3.feature_layer.layers.0.self_attn.in_proj_weight", "module.diffmodel.residual_layers.3.feature_layer.layers.0.self_attn.in_proj_bias", "module.diffmodel.residual_layers.3.feature_layer.layers.0.self_attn.out_proj.weight", "module.diffmodel.residual_layers.3.feature_layer.layers.0.self_attn.out_proj.bias", "module.diffmodel.residual_layers.3.feature_layer.layers.0.linear1.weight", "module.diffmodel.residual_layers.3.feature_layer.layers.0.linear1.bias", "module.diffmodel.residual_layers.3.feature_layer.layers.0.linear2.weight", "module.diffmodel.residual_layers.3.feature_layer.layers.0.linear2.bias", "module.diffmodel.residual_layers.3.feature_layer.layers.0.norm1.weight", "module.diffmodel.residual_layers.3.feature_layer.layers.0.norm1.bias", "module.diffmodel.residual_layers.3.feature_layer.layers.0.norm2.weight", "module.diffmodel.residual_layers.3.feature_layer.layers.0.norm2.bias", "module.diffmodel.residual_layers.4.diffusion_projection.weight", "module.diffmodel.residual_layers.4.diffusion_projection.bias", "module.diffmodel.residual_layers.4.cond_projection.weight", "module.diffmodel.residual_layers.4.cond_projection.bias", "module.diffmodel.residual_layers.4.mid_projection.weight", "module.diffmodel.residual_layers.4.mid_projection.bias", "module.diffmodel.residual_layers.4.output_projection.weight", "module.diffmodel.residual_layers.4.output_projection.bias", "module.diffmodel.residual_layers.4.time_layer.layers.0.self_attn.in_proj_weight", "module.diffmodel.residual_layers.4.time_layer.layers.0.self_attn.in_proj_bias", "module.diffmodel.residual_layers.4.time_layer.layers.0.self_attn.out_proj.weight", "module.diffmodel.residual_layers.4.time_layer.layers.0.self_attn.out_proj.bias", "module.diffmodel.residual_layers.4.time_layer.layers.0.linear1.weight", "module.diffmodel.residual_layers.4.time_layer.layers.0.linear1.bias", "module.diffmodel.residual_layers.4.time_layer.layers.0.linear2.weight", "module.diffmodel.residual_layers.4.time_layer.layers.0.linear2.bias", "module.diffmodel.residual_layers.4.time_layer.layers.0.norm1.weight", "module.diffmodel.residual_layers.4.time_layer.layers.0.norm1.bias", "module.diffmodel.residual_layers.4.time_layer.layers.0.norm2.weight", "module.diffmodel.residual_layers.4.time_layer.layers.0.norm2.bias", "module.diffmodel.residual_layers.4.feature_layer.layers.0.self_attn.in_proj_weight", "module.diffmodel.residual_layers.4.feature_layer.layers.0.self_attn.in_proj_bias", "module.diffmodel.residual_layers.4.feature_layer.layers.0.self_attn.out_proj.weight", "module.diffmodel.residual_layers.4.feature_layer.layers.0.self_attn.out_proj.bias", "module.diffmodel.residual_layers.4.feature_layer.layers.0.linear1.weight", "module.diffmodel.residual_layers.4.feature_layer.layers.0.linear1.bias", "module.diffmodel.residual_layers.4.feature_layer.layers.0.linear2.weight", "module.diffmodel.residual_layers.4.feature_layer.layers.0.linear2.bias", "module.diffmodel.residual_layers.4.feature_layer.layers.0.norm1.weight", "module.diffmodel.residual_layers.4.feature_layer.layers.0.norm1.bias", "module.diffmodel.residual_layers.4.feature_layer.layers.0.norm2.weight", "module.diffmodel.residual_layers.4.feature_layer.layers.0.norm2.bias", "module.diffmodel.residual_layers.5.diffusion_projection.weight", "module.diffmodel.residual_layers.5.diffusion_projection.bias", "module.diffmodel.residual_layers.5.cond_projection.weight", "module.diffmodel.residual_layers.5.cond_projection.bias", "module.diffmodel.residual_layers.5.mid_projection.weight", "module.diffmodel.residual_layers.5.mid_projection.bias", "module.diffmodel.residual_layers.5.output_projection.weight", "module.diffmodel.residual_layers.5.output_projection.bias", "module.diffmodel.residual_layers.5.time_layer.layers.0.self_attn.in_proj_weight", "module.diffmodel.residual_layers.5.time_layer.layers.0.self_attn.in_proj_bias", "module.diffmodel.residual_layers.5.time_layer.layers.0.self_attn.out_proj.weight", "module.diffmodel.residual_layers.5.time_layer.layers.0.self_attn.out_proj.bias", "module.diffmodel.residual_layers.5.time_layer.layers.0.linear1.weight", "module.diffmodel.residual_layers.5.time_layer.layers.0.linear1.bias", "module.diffmodel.residual_layers.5.time_layer.layers.0.linear2.weight", "module.diffmodel.residual_layers.5.time_layer.layers.0.linear2.bias", "module.diffmodel.residual_layers.5.time_layer.layers.0.norm1.weight", "module.diffmodel.residual_layers.5.time_layer.layers.0.norm1.bias", "module.diffmodel.residual_layers.5.time_layer.layers.0.norm2.weight", "module.diffmodel.residual_layers.5.time_layer.layers.0.norm2.bias", "module.diffmodel.residual_layers.5.feature_layer.layers.0.self_attn.in_proj_weight", "module.diffmodel.residual_layers.5.feature_layer.layers.0.self_attn.in_proj_bias", "module.diffmodel.residual_layers.5.feature_layer.layers.0.self_attn.out_proj.weight", "module.diffmodel.residual_layers.5.feature_layer.layers.0.self_attn.out_proj.bias", "module.diffmodel.residual_layers.5.feature_layer.layers.0.linear1.weight", "module.diffmodel.residual_layers.5.feature_layer.layers.0.linear1.bias", "module.diffmodel.residual_layers.5.feature_layer.layers.0.linear2.weight", "module.diffmodel.residual_layers.5.feature_layer.layers.0.linear2.bias", "module.diffmodel.residual_layers.5.feature_layer.layers.0.norm1.weight", "module.diffmodel.residual_layers.5.feature_layer.layers.0.norm1.bias", "module.diffmodel.residual_layers.5.feature_layer.layers.0.norm2.weight", "module.diffmodel.residual_layers.5.feature_layer.layers.0.norm2.bias", "module.diffmodel.residual_layers.6.diffusion_projection.weight", "module.diffmodel.residual_layers.6.diffusion_projection.bias", "module.diffmodel.residual_layers.6.cond_projection.weight", "module.diffmodel.residual_layers.6.cond_projection.bias", "module.diffmodel.residual_layers.6.mid_projection.weight", "module.diffmodel.residual_layers.6.mid_projection.bias", "module.diffmodel.residual_layers.6.output_projection.weight", "module.diffmodel.residual_layers.6.output_projection.bias", "module.diffmodel.residual_layers.6.time_layer.layers.0.self_attn.in_proj_weight", "module.diffmodel.residual_layers.6.time_layer.layers.0.self_attn.in_proj_bias", "module.diffmodel.residual_layers.6.time_layer.layers.0.self_attn.out_proj.weight", "module.diffmodel.residual_layers.6.time_layer.layers.0.self_attn.out_proj.bias", "module.diffmodel.residual_layers.6.time_layer.layers.0.linear1.weight", "module.diffmodel.residual_layers.6.time_layer.layers.0.linear1.bias", "module.diffmodel.residual_layers.6.time_layer.layers.0.linear2.weight", "module.diffmodel.residual_layers.6.time_layer.layers.0.linear2.bias", "module.diffmodel.residual_layers.6.time_layer.layers.0.norm1.weight", "module.diffmodel.residual_layers.6.time_layer.layers.0.norm1.bias", "module.diffmodel.residual_layers.6.time_layer.layers.0.norm2.weight", "module.diffmodel.residual_layers.6.time_layer.layers.0.norm2.bias", "module.diffmodel.residual_layers.6.feature_layer.layers.0.self_attn.in_proj_weight", "module.diffmodel.residual_layers.6.feature_layer.layers.0.self_attn.in_proj_bias", "module.diffmodel.residual_layers.6.feature_layer.layers.0.self_attn.out_proj.weight", "module.diffmodel.residual_layers.6.feature_layer.layers.0.self_attn.out_proj.bias", "module.diffmodel.residual_layers.6.feature_layer.layers.0.linear1.weight", "module.diffmodel.residual_layers.6.feature_layer.layers.0.linear1.bias", "module.diffmodel.residual_layers.6.feature_layer.layers.0.linear2.weight", "module.diffmodel.residual_layers.6.feature_layer.layers.0.linear2.bias", "module.diffmodel.residual_layers.6.feature_layer.layers.0.norm1.weight", "module.diffmodel.residual_layers.6.feature_layer.layers.0.norm1.bias", "module.diffmodel.residual_layers.6.feature_layer.layers.0.norm2.weight", "module.diffmodel.residual_layers.6.feature_layer.layers.0.norm2.bias", "module.diffmodel.residual_layers.7.diffusion_projection.weight", "module.diffmodel.residual_layers.7.diffusion_projection.bias", "module.diffmodel.residual_layers.7.cond_projection.weight", "module.diffmodel.residual_layers.7.cond_projection.bias", "module.diffmodel.residual_layers.7.mid_projection.weight", "module.diffmodel.residual_layers.7.mid_projection.bias", "module.diffmodel.residual_layers.7.output_projection.weight", "module.diffmodel.residual_layers.7.output_projection.bias", "module.diffmodel.residual_layers.7.time_layer.layers.0.self_attn.in_proj_weight", "module.diffmodel.residual_layers.7.time_layer.layers.0.self_attn.in_proj_bias", "module.diffmodel.residual_layers.7.time_layer.layers.0.self_attn.out_proj.weight", "module.diffmodel.residual_layers.7.time_layer.layers.0.self_attn.out_proj.bias", "module.diffmodel.residual_layers.7.time_layer.layers.0.linear1.weight", "module.diffmodel.residual_layers.7.time_layer.layers.0.linear1.bias", "module.diffmodel.residual_layers.7.time_layer.layers.0.linear2.weight", "module.diffmodel.residual_layers.7.time_layer.layers.0.linear2.bias", "module.diffmodel.residual_layers.7.time_layer.layers.0.norm1.weight", "module.diffmodel.residual_layers.7.time_layer.layers.0.norm1.bias", "module.diffmodel.residual_layers.7.time_layer.layers.0.norm2.weight", "module.diffmodel.residual_layers.7.time_layer.layers.0.norm2.bias", "module.diffmodel.residual_layers.7.feature_layer.layers.0.self_attn.in_proj_weight", "module.diffmodel.residual_layers.7.feature_layer.layers.0.self_attn.in_proj_bias", "module.diffmodel.residual_layers.7.feature_layer.layers.0.self_attn.out_proj.weight", "module.diffmodel.residual_layers.7.feature_layer.layers.0.self_attn.out_proj.bias", "module.diffmodel.residual_layers.7.feature_layer.layers.0.linear1.weight", "module.diffmodel.residual_layers.7.feature_layer.layers.0.linear1.bias", "module.diffmodel.residual_layers.7.feature_layer.layers.0.linear2.weight", "module.diffmodel.residual_layers.7.feature_layer.layers.0.linear2.bias", "module.diffmodel.residual_layers.7.feature_layer.layers.0.norm1.weight", "module.diffmodel.residual_layers.7.feature_layer.layers.0.norm1.bias", "module.diffmodel.residual_layers.7.feature_layer.layers.0.norm2.weight", "module.diffmodel.residual_layers.7.feature_layer.layers.0.norm2.bias", "module.diffmodel.residual_layers.8.diffusion_projection.weight", "module.diffmodel.residual_layers.8.diffusion_projection.bias", "module.diffmodel.residual_layers.8.cond_projection.weight", "module.diffmodel.residual_layers.8.cond_projection.bias", "module.diffmodel.residual_layers.8.mid_projection.weight", "module.diffmodel.residual_layers.8.mid_projection.bias", "module.diffmodel.residual_layers.8.output_projection.weight", "module.diffmodel.residual_layers.8.output_projection.bias", "module.diffmodel.residual_layers.8.time_layer.layers.0.self_attn.in_proj_weight", "module.diffmodel.residual_layers.8.time_layer.layers.0.self_attn.in_proj_bias", "module.diffmodel.residual_layers.8.time_layer.layers.0.self_attn.out_proj.weight", "module.diffmodel.residual_layers.8.time_layer.layers.0.self_attn.out_proj.bias", "module.diffmodel.residual_layers.8.time_layer.layers.0.linear1.weight", "module.diffmodel.residual_layers.8.time_layer.layers.0.linear1.bias", "module.diffmodel.residual_layers.8.time_layer.layers.0.linear2.weight", "module.diffmodel.residual_layers.8.time_layer.layers.0.linear2.bias", "module.diffmodel.residual_layers.8.time_layer.layers.0.norm1.weight", "module.diffmodel.residual_layers.8.time_layer.layers.0.norm1.bias", "module.diffmodel.residual_layers.8.time_layer.layers.0.norm2.weight", "module.diffmodel.residual_layers.8.time_layer.layers.0.norm2.bias", "module.diffmodel.residual_layers.8.feature_layer.layers.0.self_attn.in_proj_weight", "module.diffmodel.residual_layers.8.feature_layer.layers.0.self_attn.in_proj_bias", "module.diffmodel.residual_layers.8.feature_layer.layers.0.self_attn.out_proj.weight", "module.diffmodel.residual_layers.8.feature_layer.layers.0.self_attn.out_proj.bias", "module.diffmodel.residual_layers.8.feature_layer.layers.0.linear1.weight", "module.diffmodel.residual_layers.8.feature_layer.layers.0.linear1.bias", "module.diffmodel.residual_layers.8.feature_layer.layers.0.linear2.weight", "module.diffmodel.residual_layers.8.feature_layer.layers.0.linear2.bias", "module.diffmodel.residual_layers.8.feature_layer.layers.0.norm1.weight", "module.diffmodel.residual_layers.8.feature_layer.layers.0.norm1.bias", "module.diffmodel.residual_layers.8.feature_layer.layers.0.norm2.weight", "module.diffmodel.residual_layers.8.feature_layer.layers.0.norm2.bias", "module.diffmodel.residual_layers.9.diffusion_projection.weight", "module.diffmodel.residual_layers.9.diffusion_projection.bias", "module.diffmodel.residual_layers.9.cond_projection.weight", "module.diffmodel.residual_layers.9.cond_projection.bias", "module.diffmodel.residual_layers.9.mid_projection.weight", "module.diffmodel.residual_layers.9.mid_projection.bias", "module.diffmodel.residual_layers.9.output_projection.weight", "module.diffmodel.residual_layers.9.output_projection.bias", "module.diffmodel.residual_layers.9.time_layer.layers.0.self_attn.in_proj_weight", "module.diffmodel.residual_layers.9.time_layer.layers.0.self_attn.in_proj_bias", "module.diffmodel.residual_layers.9.time_layer.layers.0.self_attn.out_proj.weight", "module.diffmodel.residual_layers.9.time_layer.layers.0.self_attn.out_proj.bias", "module.diffmodel.residual_layers.9.time_layer.layers.0.linear1.weight", "module.diffmodel.residual_layers.9.time_layer.layers.0.linear1.bias", "module.diffmodel.residual_layers.9.time_layer.layers.0.linear2.weight", "module.diffmodel.residual_layers.9.time_layer.layers.0.linear2.bias", "module.diffmodel.residual_layers.9.time_layer.layers.0.norm1.weight", "module.diffmodel.residual_layers.9.time_layer.layers.0.norm1.bias", "module.diffmodel.residual_layers.9.time_layer.layers.0.norm2.weight", "module.diffmodel.residual_layers.9.time_layer.layers.0.norm2.bias", "module.diffmodel.residual_layers.9.feature_layer.layers.0.self_attn.in_proj_weight", "module.diffmodel.residual_layers.9.feature_layer.layers.0.self_attn.in_proj_bias", "module.diffmodel.residual_layers.9.feature_layer.layers.0.self_attn.out_proj.weight", "module.diffmodel.residual_layers.9.feature_layer.layers.0.self_attn.out_proj.bias", "module.diffmodel.residual_layers.9.feature_layer.layers.0.linear1.weight", "module.diffmodel.residual_layers.9.feature_layer.layers.0.linear1.bias", "module.diffmodel.residual_layers.9.feature_layer.layers.0.linear2.weight", "module.diffmodel.residual_layers.9.feature_layer.layers.0.linear2.bias", "module.diffmodel.residual_layers.9.feature_layer.layers.0.norm1.weight", "module.diffmodel.residual_layers.9.feature_layer.layers.0.norm1.bias", "module.diffmodel.residual_layers.9.feature_layer.layers.0.norm2.weight", "module.diffmodel.residual_layers.9.feature_layer.layers.0.norm2.bias", "module.diffmodel.residual_layers.10.diffusion_projection.weight", "module.diffmodel.residual_layers.10.diffusion_projection.bias", "module.diffmodel.residual_layers.10.cond_projection.weight", "module.diffmodel.residual_layers.10.cond_projection.bias", "module.diffmodel.residual_layers.10.mid_projection.weight", "module.diffmodel.residual_layers.10.mid_projection.bias", "module.diffmodel.residual_layers.10.output_projection.weight", "module.diffmodel.residual_layers.10.output_projection.bias", "module.diffmodel.residual_layers.10.time_layer.layers.0.self_attn.in_proj_weight", "module.diffmodel.residual_layers.10.time_layer.layers.0.self_attn.in_proj_bias", "module.diffmodel.residual_layers.10.time_layer.layers.0.self_attn.out_proj.weight", "module.diffmodel.residual_layers.10.time_layer.layers.0.self_attn.out_proj.bias", "module.diffmodel.residual_layers.10.time_layer.layers.0.linear1.weight", "module.diffmodel.residual_layers.10.time_layer.layers.0.linear1.bias", "module.diffmodel.residual_layers.10.time_layer.layers.0.linear2.weight", "module.diffmodel.residual_layers.10.time_layer.layers.0.linear2.bias", "module.diffmodel.residual_layers.10.time_layer.layers.0.norm1.weight", "module.diffmodel.residual_layers.10.time_layer.layers.0.norm1.bias", "module.diffmodel.residual_layers.10.time_layer.layers.0.norm2.weight", "module.diffmodel.residual_layers.10.time_layer.layers.0.norm2.bias", "module.diffmodel.residual_layers.10.feature_layer.layers.0.self_attn.in_proj_weight", "module.diffmodel.residual_layers.10.feature_layer.layers.0.self_attn.in_proj_bias", "module.diffmodel.residual_layers.10.feature_layer.layers.0.self_attn.out_proj.weight", "module.diffmodel.residual_layers.10.feature_layer.layers.0.self_attn.out_proj.bias", "module.diffmodel.residual_layers.10.feature_layer.layers.0.linear1.weight", "module.diffmodel.residual_layers.10.feature_layer.layers.0.linear1.bias", "module.diffmodel.residual_layers.10.feature_layer.layers.0.linear2.weight", "module.diffmodel.residual_layers.10.feature_layer.layers.0.linear2.bias", "module.diffmodel.residual_layers.10.feature_layer.layers.0.norm1.weight", "module.diffmodel.residual_layers.10.feature_layer.layers.0.norm1.bias", "module.diffmodel.residual_layers.10.feature_layer.layers.0.norm2.weight", "module.diffmodel.residual_layers.10.feature_layer.layers.0.norm2.bias", "module.diffmodel.residual_layers.11.diffusion_projection.weight", "module.diffmodel.residual_layers.11.diffusion_projection.bias", "module.diffmodel.residual_layers.11.cond_projection.weight", "module.diffmodel.residual_layers.11.cond_projection.bias", "module.diffmodel.residual_layers.11.mid_projection.weight", "module.diffmodel.residual_layers.11.mid_projection.bias", "module.diffmodel.residual_layers.11.output_projection.weight", "module.diffmodel.residual_layers.11.output_projection.bias", "module.diffmodel.residual_layers.11.time_layer.layers.0.self_attn.in_proj_weight", "module.diffmodel.residual_layers.11.time_layer.layers.0.self_attn.in_proj_bias", "module.diffmodel.residual_layers.11.time_layer.layers.0.self_attn.out_proj.weight", "module.diffmodel.residual_layers.11.time_layer.layers.0.self_attn.out_proj.bias", "module.diffmodel.residual_layers.11.time_layer.layers.0.linear1.weight", "module.diffmodel.residual_layers.11.time_layer.layers.0.linear1.bias", "module.diffmodel.residual_layers.11.time_layer.layers.0.linear2.weight", "module.diffmodel.residual_layers.11.time_layer.layers.0.linear2.bias", "module.diffmodel.residual_layers.11.time_layer.layers.0.norm1.weight", "module.diffmodel.residual_layers.11.time_layer.layers.0.norm1.bias", "module.diffmodel.residual_layers.11.time_layer.layers.0.norm2.weight", "module.diffmodel.residual_layers.11.time_layer.layers.0.norm2.bias", "module.diffmodel.residual_layers.11.feature_layer.layers.0.self_attn.in_proj_weight", "module.diffmodel.residual_layers.11.feature_layer.layers.0.self_attn.in_proj_bias", "module.diffmodel.residual_layers.11.feature_layer.layers.0.self_attn.out_proj.weight", "module.diffmodel.residual_layers.11.feature_layer.layers.0.self_attn.out_proj.bias", "module.diffmodel.residual_layers.11.feature_layer.layers.0.linear1.weight", "module.diffmodel.residual_layers.11.feature_layer.layers.0.linear1.bias", "module.diffmodel.residual_layers.11.feature_layer.layers.0.linear2.weight", "module.diffmodel.residual_layers.11.feature_layer.layers.0.linear2.bias", "module.diffmodel.residual_layers.11.feature_layer.layers.0.norm1.weight", "module.diffmodel.residual_layers.11.feature_layer.layers.0.norm1.bias", "module.diffmodel.residual_layers.11.feature_layer.layers.0.norm2.weight", "module.diffmodel.residual_layers.11.feature_layer.layers.0.norm2.bias".

[yhd-ai]

And I wrote codes to make the keys aligned.
BTW, in File "main_tcd_h36m.py",
output = model_s.module.evaluate(s, nsample) seems to be output = model_s.evaluate(s, nsample)
cuz 'ModelMain' object has no attribute 'module'

[yhd-ai]

Learned a lot