ValueError: loaded state dict contains a parameter group that doesn't match the size of optimizer's group

Hi

I killed my training and launch a resume from the last checkpoint.
I did that before without issue (to add more images in the training) but this time I got this error:

File "E:\kohya\venv\lib\site-packages\bitsandbytes\optim\optimizer.py", line 149, in load_state_dict
raise ValueError(
ValueError: loaded state dict contains a parameter group that doesn't match the size of optimizer's group

I don't understand why.

Start by reproducing resume training from the last checkpoint and inspect the traceback at bitsandbytes/optim/optimizer.py line 149. Compare the saved optimizer parameter groups with those created for the resumed run, then document the cause and verify that resuming succeeds without the mismatch.

Written by the indexing model from the issue text.