Fabian Isensee
Fabian Isensee
(Reading helps, my bad! I didn't see you posted the error message. Apologies.) The software simply does not support Ampere. That is probably unfixable without recreating each container using a...
Hi @vincenzoml , the problem is not the dockerfile. The problem is that in order to recreate the docker with a newer NGC base you need to have access to...
I believe this is indeed something you could to try and bug nvidia with :-D AFAIK you cannot change the base image of a container after it was created (I...
Yeah this is just a problem with hiddenlayer and pytorch 1.12. You can ignore this as this will not interfere with your training. It will just be unable to create...
Has someone been able to get hiddenlayer to work with pytorch 1.12?
Remember to restore the geometry (spacing etc) when exporting them ;-)
Hi, you need to save each 3D image from the 4D sequence in a separate file. Then you can predict them with nnU-Net. Best, Fabian
Use `nnUNetTrainerV2_2_noMirror` For more information about each trainer I recommend you take a look at the code. Best, Fabian
And beware of the `all` fold. I prefer running cross-validation to get an accurate validation score (whcih all does NOR provide)
your `threadpoolctl` seems to be broken. Have you tried reinstalling it?