search

yeatmanlab / AFQ-Browser

Browser-based visualization tools for AFQ results
BSD 3-Clause "New" or "Revised" License
33 stars 15 forks source link

faPlotLength #184

Closed arokem closed 6 years ago

arokem commented 6 years ago

Currently, the variable faPlotLength holds the expected length of the plots of the metrics as one single number. However, in Tracula, different tracts may have different lengths. As an example, this code checks that all the streamlines have that same length, so will need to be refactored to point to a list (?) of lengths.

richford commented 6 years ago

Currently we require that all fibers be the same length, but for starts there's no reason that it has to be faPlotLength. So #188 removes that dependence. But this doesn't necessarily solve this issue.

Tracula only has one central core fiber right? In that case, then #188 fixes this since each bundle is allowed to have a different length.

But for performance reasons and to make brushing easier to implement, it is really nice to require that each fiber in a bundle is the same length. So if we need to have multiple fibers of varying length within a single bundle, then I'll have to think more about how to implement this.

arokem commented 6 years ago

I think it's fine to assume that all the fibers within a bundle are resampled to have the same length, allowing for different bundles to have different lengths. That means #188 solves this, right?

richford commented 6 years ago

Yep. #188 should fix this. I still haven't implemented the differentiation between core fiber and other fibers. But #188 should fix this specific issue.

richford commented 6 years ago

Resolved by #188.