You signed in with another tab or window. Reload to refresh your session.You signed out in another tab or window. Reload to refresh your session.You switched accounts on another tab or window. Reload to refresh your session.Dismiss alert
We are already enabling GPU-accelerated tractography via the GPUStreamlines project.
I wonder if we could push this even further by using cudipy, e.g., for registration. It looks like maybe cudipy directly translates the API of syn registration, for example: https://github.com/dipy/cudipy/blob/master/cudipy/align/imwarp.py, in a way that we could ducktype the inputs and outputs as needed.
Might also be a thing that needs to be done upstream in dipy itself, though.
The text was updated successfully, but these errors were encountered:
We are already enabling GPU-accelerated tractography via the GPUStreamlines project.
I wonder if we could push this even further by using cudipy, e.g., for registration. It looks like maybe cudipy directly translates the API of syn registration, for example: https://github.com/dipy/cudipy/blob/master/cudipy/align/imwarp.py, in a way that we could ducktype the inputs and outputs as needed.
Might also be a thing that needs to be done upstream in dipy itself, though.
The text was updated successfully, but these errors were encountered: