Open zhaoyong-li opened 4 months ago
same problem! I also try to run inference.py by Real-DRCT-GAN_MSE_Model, but get the RuntimError:
Traceback (most recent call last):
File "inference.py", line 111, in
Please tell me how can i solve this issue?
Hello guys, I had the same issue in a similar model, and solved it by checking that the weights i tried to load for the model correspond to the architecture of the model in the inference.py file. This error means that the layers expected by the weights differ from the architecture of the model initialized in the inference.py file. According to the default argument model_path in inference.py file, model_weights like DRTCT-L.pth are expected, which ones are you trying to use ? I would have tried to use these ones : https://drive.google.com/file/d/1bVxvA6QFbne2se0CQJ-jyHFy94UOi3h5/view If they don't work, we either need to find other weights corresponding to the architecture, or changing the architecture in the model initialization in the inference.py file. Let me know if it worked :)
Hello, how can I solve the above problem, I am also having the same problem with net_g_latest.pth and DRCT-L_X4.pth, using net_g_latest (MSEModel).pth tests normally
hi, thank you for providing such great work! But I have a question when using Real-DRCT-GAN_MSE_Model for inference, is it not possible to use the inference file directly for this model weight? When I try, the following problem occurs : "RuntimeError: Error(s) in loading state_dict for DRCT". I don't know if there is something wrong with my operation and would appreciate your answer!