Open seedlessbuns opened 1 year ago
I also got this error, changed network.load_state_dict(load_net_clean, strict=strict) to network.load_state_dict(load_net_clean, strict=False) in codes/trainer/basem_model.py
seems to be training now, but no idea what effect this will have on the end model
I also got this error, changed network.load_state_dict(load_net_clean, strict=strict) to network.load_state_dict(load_net_clean, strict=False) in codes/trainer/basem_model.py
seems to be training now, but no idea what effect this will have on the end model
Tried using the model that I just trained with this change, and it looks like it doesn't work.
I guess this repo is broken in its current state.
I am getting the same errors, no clue as to why. I already tried reinstalling and going over the process again. It is a problem with torch, but I am unsure what the problem is exactly.
same.
Shout out to @FurkanGozukara he has an updated repository with the fix for this. Short answer. Update Transformers to 4.28.1 pip install transformers==4.28.1
Shout out to @FurkanGozukara he has an updated repository with the fix for this. Short answer. Update Transformers to 4.28.1 pip install transformers==4.28.1
thank you so much
here the updated repo. i try to keep my tutorials working :)
also here my tutorial
After starting training from the GUI and from CLI I get this error. Not sure what is going on, could I have the wrong autoregressive.pth? I don't see how this is the case, but it seems like something is not formatted/reformatted correctly from my very basic understanding of the State Dict.