-
Notifications
You must be signed in to change notification settings - Fork 67
Description
Describe the bug
When running DeepMedic in a container on a cluster with singularity, I get the following error :
Starting DeepMedic Segmentation.
sh: /work/CaPTk/bin/deepMedicInference: Is a directory
DeepMedic exited with code !=0.
Something went wrong when performing skull-stripping using DeepMedic, please re-try or contact sofware@cbica.upenn.edu.
I get this error when running DeepMedic by itself or through the BraTSPipeline.
To Reproduce
Steps to reproduce the behavior:
- Pull container from dockerhub with singularity pull docker://cbica/captk:{version}
- Run container (interactive or not) with command /work/CaPTk/bin/DeepMedic -md /work/CaPTk/bin/../data//deepMedic/saved_models/skullStripping/ -i /output//T1_to_SRI.nii.gz,/output//T1CE_to_SRI.nii.gz,/output//T2_to_SRI.nii.gz,/output//FL_to_SRI.nii.gz -o /output//dmOut_skull/brainMask_SRI.nii.gz
- See error
Expected behavior
DeepMedic doesn't get stuck and gives an output
Screenshots
If applicable, add screenshots to help explain your problem.
CaPTk Version
azure-latest, but also in latest and all previous containers
HPC:
fedora 8.6
Additional context
In the BraTSPipeline everything before the DeepMedic segmentation runs without issue as intended.