Vibhu Jawa

> For 3) would could ask folks like @VibhuJawa @ayushdg and @randerzander for their input in building their defaults as highlighted options More than happy to help on this ....

> Is this a fix for cugraph or pytorch? Unkown honestly.

@ryantwolf , Can you confirm the Pytorch/cugraph version you were seeing for this issue ?

> How about we add a new method `sample_multilayer_neighbors` that directly returns a list of blocks, and then in `NeighborSampler.sample_blocks` test if the graph storage has this method? This sounds...

I just wanted to share an update on this. With the current work in progress [minimal](https://github.com/alteryx/featuretools/compare/main...VibhuJawa:cudf_poc) changes to support `cudf`, the [loan repayment benchmark](https://gist.github.com/VibhuJawa/d2a9872a4fc73dcace9e35a2b065eb05) shared [here](https://github.com/alteryx/featuretools/issues/1213#issuecomment-722673074) without the `cutoff-times` bit,...

> We are pausing on this until GitHub actions supports GPU instances One option to use GPU instancess with Github actions is to use self hosted GPU runners. The cuDF...

@quasiben , Wondering if we have opinion on this ? Happy to do a PR here to make life easier for folks like me. CC: @alexbarghi-nv, @jnke2016 who have seen...

> @VibhuJawa apologies, for not responding earlier. When I tried this the workflow took +45 minutes at which point I killed it. Is this your experience ? Nope, It takes...

> I don't know if this is something we are ready to do yet for the general use case, [JIT-Unspill for CuPy is still very slow](https://github.com/rapidsai/dask-cuda/issues/840#issuecomment-1143288209). > > cc @madsbk...

> I don't think that is how `rebalance` is intended to be used as the first argument is `comm` not a Dask object. I thought its supposed to a `futures`...