Fabian Isensee
Fabian Isensee
Thanks a lot for this work. I am happy to add this as a github action. To make sure people can see that you contributed the test please make a...
I took a look at the commit history. The only thing I changed that day (July 26th) was switching from torch.inference_mode back to torch.no_grad
Thanks! Christmas time is super busy and I am not getting anything done at the moment. I have a private PC that I can use for debugging but this will...
Has anyone found the problem? I still do not have a Windows setup to check what's going on and why this fails
I just pushed a change reverting back to torch.no_grad. This alleviates the problem. Cloning potentially large tensors is not a good idea. I have an idea how to fix this...
Thanks for this PR. I am not a big fan of this to be honest because we take something that is well structured and abstracted (labelmanager is in charge) and...
Hey, so I would like to close this PR because it breaks with our existing design patterns to work around a problem caused by a single challenge platform. I suggest...
The mmap_mode='r' is there for a reason. On some systems read performance can tank very hard if it's not enabled. From your comments it appears that you are all using...
OK I guess we need to temporarily remove the memmap part for Windows. This is now implemented on master. Since this is related to blosc2 I will create a new...
You still are on an old version of nnU-Net that doesn't use blosc2