maggie2209

@constantinulrich Are you aware of anything else I could try to fix this?

I just tried this and yes, the same error appears without the pretrained weights.

Thank you, here is the plan: [plans.json](https://github.com/MIC-DKFZ/nnUNet/files/15486205/plans.json)

Thank you for checking. Yes, the pretraining works fine, I just tried it again to be sure, so I assume there are no issues with cuda or pytorch. I don't...

@constantinulrich My supervisor is happy for me to share the finetuning data with you. Would that be helpful?

Update in case anyone encounters the same problem: I was able to fix it by installing a completely new nnunet environment with `virtualenv` and re-running the processing in there.