optional Output of label probability maps by nnUnet
- Dominant language
- Python
- Stars
- 13
- Forks
- 11
- Avg merge
- 1m
- Merged PRs (30d)
- 1
Description
### A short summary of what you would like to see in BIBSnet.
I was wondering if an additional output could be generated if requested, which would be the voxel-wise probabilities of each label. I think that you can get the raw probability maps by enabling the softmax export: e.g.
nnUNet_predict \
-i INPUT_FOLDER \
-o OUTPUT_FOLDER \
-t TASK_ID \
-m 3d_fullres \
--save_npz
This might be useful to some folk who might want to differentiate between different WM probabilities, for example.
### Do you have any interest in helping implement the feature?
Yes, but I would need guidance
### Add any additional information or context about the request here.
_No response_
Contributor guide
No contributing guide indexed for this repository
Assessment
This issue has not been assessed yet.