Fabian Isensee
Fabian Isensee
Hey, it's hard to say what could be going on. My recommendations would be the following: - try the `nnUNetTrainerV2_Loss_DiceCE_noSmooth` trainer - increase the batch size - manually set batch_dice...
- the plot you are showing does not fit the log - your epoch time is abysmal hinting at some problem with your setup. This should be fixed first because...
nnUNet_determine_postprocessing should do the trick ;-)
This has been described multiple times already. Please use the search function. Read this: https://github.com/MIC-DKFZ/nnUNet/blob/master/documentation/extending_nnunet.md In principle, each change requires a separate trainer class. Changing epochs works like this: ```...
There was an old version of nnunet where this could happen (like years old, nothing recent) but we have since fixed this problem. On 3D datasets, the 3D configurations always...
@chrisrapson correct! Running all nnU-Net configurations is only really needed for users who don't know much about it (yet). Once you get a feeling for the datasets you will know...
Hey there, converting file formats to something that is interpretable by nnU-Net is not within the scopeof nnU-Net (how could it be?), so I cannot help you here. I will...
Yes it seems like your images were not registered. The images do not have the same geometry
Yeah what @BigPandaCPU says. Unless you know RAM is not an issue
Guys you are using expensive machines so you should also know how to monitor their resource usage. If the CPU is full it is full and there is unfortunately not...