FAQ

Numerous errors can happen during the pipeline for several reasons such as hardware limitation, bad data, bad configuration files, ...

Here are some tips and tricks about the errors that were already encountered:

During the installation process, I get the following error: ./remesher/cmdremesher/cmdremesher: error while loading shared libraries: libgsl.so.0: cannot open shared object file: No such file or directory

The issue was resolved by running the following command prior to running the pipeline: sudo ln -s find /usr -name libgsl.so{,.0} see https://github.com/ins-amu/scripts/issues/67

I have set up a VM with the processed data on the host computer, but Freesurfer fails

Freesurfer does not support working on VMs with data on the host machine. You should set the freesurfer directory in the guest hard drive

The pipeline suddenly fails in the middle of the mrconvert command

Check if you have enough RAM memory. For instance, HCP dataset uncompressed for dMRI are huge (8 Gb), if they cannot hold in memory the pipeline will fail.

I have two big regions when checking the region mapping for the surface, which one should I delete.

Probably none of them, they should be connected. Try to rerun the pipeline by changing the magic value. Otherwise the region mapping can be corrected by hand, for example using anatomist

After the pipeline fails, can I rerun the pipeline?

Probably yes, it should start back from the place it fails. However if that does not work, you should check in the main surface the test condition where you were, and delete the file for that condition

I have Windows, can I use the pipeline?

No, unless you set up a VM