Change approach for official submissions accepting mha files to simply convert those to .nii.gz at an intermediate location and the run predictions as normal, and then finally converting the .nii.gz predictions back to the expected format.
Checklist
[x] Merged latest master
[ ] Updated version number in README.md
[ ] Added changes to changelog.md
[ ] Updated version number in setup.py
[ ] (only when updating dataset) Ran annotation.import to completion
[ ] (only when updating dataset) Updated Surface Dice tolerances in labels.py (execute sample_segmentations.py, and
after that compute_tolerances.py. Then put the new values in. Do not re-use sampled segmentations from prior dataset versions!)
Handle MHA Inputs By Converting to Nifti
Description
Change approach for official submissions accepting mha files to simply convert those to .nii.gz at an intermediate location and the run predictions as normal, and then finally converting the .nii.gz predictions back to the expected format.
Checklist
masterREADME.mdchangelog.mdsetup.pyannotation.importto completionlabels.py(executesample_segmentations.py, and after thatcompute_tolerances.py. Then put the new values in. Do not re-use sampled segmentations from prior dataset versions!)