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
I think we basically want to be pretty lightweight here: much more serious people have done much more serious work for regularized maximum likelihood image reconstruction for EHT, radio VLBI, and ALMA. In my view, we are not trying to compete with that: but to provide a user-friendly, lightweight Jax interface to our own dLux and interferometry code for (at least to start with) a restricted set of problems.
The much more fully-featured PyTorch package and team Million Points of Light might be a good source of inspiration and collaboration.
The text was updated successfully, but these errors were encountered:
I think we basically want to be pretty lightweight here: much more serious people have done much more serious work for regularized maximum likelihood image reconstruction for EHT, radio VLBI, and ALMA. In my view, we are not trying to compete with that: but to provide a user-friendly, lightweight Jax interface to our own dLux and interferometry code for (at least to start with) a restricted set of problems.
The much more fully-featured PyTorch package and team Million Points of Light might be a good source of inspiration and collaboration.
The text was updated successfully, but these errors were encountered: