Daniel Coquelin

there is a version of this implemented in heat now however it is based on nonblocking hooks but it still does traditional synchronization SGD. that is not traditional ASGD. there...

I am not 100% sure, but it is likely. i think this can be closed for now under this assumption. If we see something in the docs then we can...

i am working on tiling on my QR branch

small update on this one. the #520 PR has a new tiling class. theoretically, this could be modified to cope with this by only sending partial tiles. although it may...

While this is something we would love to have, it is something that we are not currently pursuing. Firstly, the main problem is that we do not have a sparse...

what branch are you on? can you try this on the `features/436-SVD-block-diag` branch? there are some features there which are not fully functional, but there were some changes to the...

another container to try: https://catalog.ngc.nvidia.com/orgs/nvidia/containers/pytorch

rerun tests

i am pretty sure that this is caused by the assumption if the index is a split `DNDarray`, it is assumed to be non-repeating. when the indices are gathered they...

I think that removing the division entirely is not the best way to go about this. I think that the reason that it is not taken up is because it...