Ole Johannsen

Dear VitalityWu, sorry for the late reply. Did you already solve the issue? I need some more details to help you. At which point does this problem occur? What data...

Hey! Did you manage to solve the issue by now? Have you tried training without pretrained weights? Cheers, Ole

Unfortunately, this is not possible.

Hello, Did you manage to resolve the problem or do you need further assistance? Otherwise, I will close the issue. Cheers Ole

Dear Alex, thanks for the feedback. I will bring this idea to Fabians attention. Best regards, Ole

Is it possible that the image in question has a different shape than the other ones, i.e. more channels?

Hello, sorry for the late response. Its this question still relevant or did you manage to solve it?

The number given specifies the fold that is run and should be in [0,1,2,3,4] or 'all'. nnUnetv2 automatically performs 5 fold cross-validation. If the number is bigger than 4 a...

You can find information about bow to run inference in the documentation: https://github.com/MIC-DKFZ/nnUNet/blob/master/documentation/how_to_use_nnunet.md#run-inference

does following this suggestion help: https://github.com/pytorch/pytorch/issues/37377#issuecomment-629530272?