Matthias Fey

If you want to contribute to PyG, you need to fork this repo, push the changes to your local fork, and then submit a PR to PyG.

Do you mind sharing a small code snippet to understand this better? If your input is a regular grid, there is no real point in using a GNN.

I am running ```python data = Data( z=torch.randint(1, 10, (20, )), pos=torch.randn(20, 3), ) model = SchNet( hidden_channels=16, num_filters=16, num_interactions=2, num_gaussians=10, cutoff=6.0, dipole=True, atomref=torch.randn(100, 1) if use_atomref else None, )...

I somehow missed this issue, sorry. This should be resolved on latest master already.

Hey @iakovosevdaimon. This is a welcome contribution. Do you have a place to upload the raw data somewhere? Depending on the graph size, we might also wanna us `OnDiskDataset` rather...

Ok, do you also want to contribute the code to it?

Yes :)

I can't think of an example on top of my head, but usually it is not hart to convert a node classification task to a node regression task. It usually...

Yeah, this is well supported within `pyg-lib`, just not really exposed in PyG. This is not super high on my TODO, so I would appreciate some help on this one...

Can you share more information about your PyTorch version, and what the installation log look like if you run via `pip install --verbose`?