Currently dealing with a very large parcellation image (voxel-wise, specifically only looking at the wm/gm interface) as input for tck2connectome in an attempt to obtain a structural connectivity matrix from a whole-brain tracrtography. I’m having an issue with the allocation of memory (the matrix created is far too big). Any insight on reducing this effectively or other methods on how to conduct a voxel connectivity matrix would be much appreciated.
Cheers,
Ricardo
I’m having an issue with the allocation of memory (the matrix created is far too big).
Yeah, that’ll happen 
The issue is already listed on GitHub here, and was discussed previously on the forum here. While it’s certainly technically possible, it’s not a trivial change, and unfortunately without a necessity for use myself I am very unlikely to prioritize it over other things. It would however be a good little project for e.g. a computer science undergrad, if anybody interested comes across such an opportunity.