Julien Cohen-Adad

thank you for the thorough investigation! this is elegant detective work

A quick and easy way to solve this would be to catch an empty input seg and throw an error if that happens

> We don't pass `fname_src_label` or `fname_dest_label` at all. Indeed, in the `-ref template` case, the label-based alignment is done earlier in the code: https://github.com/spinalcordtoolbox/spinalcordtoolbox/blob/7fd2ea718751dd858840c3823c0830a910d9777c/spinalcordtoolbox/scripts/sct_register_to_template.py#L600 > So, it might just...

> Specifically, I believe this is referring to the fact that sct_deepseg_lesion can be run on images of multiple orientations: to clarify, `-c` specifically refers to the contrast, and _sometimes_...

> I thought that for sct_deepseg_lesion specifically, we had used -c in a different way than usual. (Since we have both t2 and t2_ax options, which are used to pick...

Here is an example: https://github.com/sct-pipeline/bavaria-quebec/blob/d5c7f5424facf724551afdc40be2ab8310891bcf/source/pipeline/brain/process_brain_data_t1.sh needs to be curated, but i'm dumping it there so at least we have something

thank you for catching that @kousu

Related to https://github.com/sct-pipeline/contrast-agnostic-softseg-spinalcord

@charleygros while doing the refactoring it would be a good opportunity to remove this dependency: https://github.com/neuropoly/spinalcordtoolbox/pull/2739/commits/f8faaab439815ede36526b914e3d1512e71b8a84

> However, after some time the python script (here sct_deepseg_sc.py) will have no limitation anymore on the number of CPUs to use, resulting on all the CPUs of the machine...