Justin Wang
Justin Wang
Issue: Pull request shows commits that have already been brought over
Tract length matrices differ depending on probtrackx implementation (GPU, GPU batch, CPU, CPU batch)
There is only ~ 0.66 correlation between tract length matrix outputs (for the same input data) of the pipeline's various probtrackx options: GPU, GPU batch, CPU, CPU batch. This suggests...
1. update main branch with ADNI reparcellation tool 2. update wiki documentation re: reparcellating only on subject run with a version of the pipeline that is reparcellation-capable (post-reparcellation tool commit)....
Currently, toggling the GPU flag off in init-vars may produce errors due to commands for which we have not provided CPU alternatives (e.g. eddy_cuda in bb_eddy_wrap). Other potential errors have...
currently quits if sc missing line must be in try block: https://github.com/McIntosh-Lab/tvb-ukbb/blob/5f59271d40b24c006a4f966250e6d5571d7a231e/tvb_bb_QC/new_IDP_gen.py#L164