Hatem Helal

I can take a look at this: as a first attempt I'll try to relax the check for `edge_index.dtype == torch.long` and add some additional tests in `test/nn/conv/test_message_passing.py` and see...

See #5281 for a proposal to relax the type assertion in the message passing interface. This patch effectively lets the execution backend decide which integer types to support when executing...

@LiuHaolan that would be amazing if you wanted to get this going as it was going to take me a few weeks before I can work on this. Happy to...

There are probably many other ways to achieve this but I think I had in mind that the `ChunkedDataset` interface would be responsible for the chunking logic. The `process_chunk` method...

I think targeting a pyg version of `IterableDataset` makes a lot of sense. I really like the feature of applying a `pre_transform` as a static preprocessing that only needs to...

I think some of the issues might go away by upgrading to 0.4 JAX (or at least there will probably be new issues). I think there should be some cmath...

I had the idea to introduce a function decorator that will run an operation twice: 1. a baseline run with all the floating point inputs promoted to fp64 2. a...

We should look at this action: https://github.com/benchmark-action/github-action-benchmark

I put together #110 which contains just the function decorator idea: the problem I haven't solved is how to inject it into the desired place within a larger compute graph....

@AlexanderMath is there any reason to prefer using the `6-31G` basis by default? I'd like to make `sto-3g` the default.