Firstly thanks for the great jobs. However, I encountered some issues during validation and testing. I only trained using the BTCV dataset. After training, I faced problems when using the trained epoch_xx.pth for validation.
RuntimeError: Error(s) in loading state_dict for Universal_model:
Unexpected key(s) in state_dict: "", "0.weight", "0.bias", "2.weight", "2.bias", "3.weight", "3.bias", "weight", "bias".
When I used backbone=’swinunetr’ and trained './out/swinunetr/epoch_xx.pth’, I encountered an error.
KeyError: ‘Key label of transform Invertd was missing in the data and allow_missing_keys==False.’
when trying to run 'pred_pseudo.py’. How can I resolve this issue?
Firstly thanks for the great jobs. However, I encountered some issues during validation and testing. I only trained using the BTCV dataset. After training, I faced problems when using the trained epoch_xx.pth for validation.
RuntimeError: Error(s) in loading state_dict for Universal_model: Unexpected key(s) in state_dict: "", "0.weight", "0.bias", "2.weight", "2.bias", "3.weight", "3.bias", "weight", "bias".![image](https://github.com/ljwztc/CLIP-Driven-Universal-Model/assets/65030311/8de22ef2-4be5-43eb-8f23-142d9cbb6ad7)
When I used backbone=’swinunetr’ and trained './out/swinunetr/epoch_xx.pth’, I encountered an error.
KeyError: ‘Key label of transform Invertd was missing in the data and allow_missing_keys==False.’
when trying to run 'pred_pseudo.py’. How can I resolve this issue?