Matthias Fey

Thanks for reporting. This issue is introduced in PyG 2.0, and it is now fixed in PyG master, see [here](https://github.com/pyg-team/pytorch_geometric/commit/724ec4cb088d68e3fbba7c4152073b6a30c9f6ab). You can either apply the change locally, or install PyG...

If I understand you correctly, you generate DeepWalk features separately for each graph. Our matching procedure is based on finding correspondences based on locally similar neighborhoods (which is not given...

Thanks a lot for reporting. While I hadn't have time to reproduce this myself, my first guess is that this is related to Windows. See [here](https://github.com/rusty1s/pytorch_sparse/pull/198) for more information. We...

Amazing. Can you send a PR for `torch-cluster` as well? This would be great!

Hi and thanks for your interest! Sorry, I missed that issue. I haven't got around to clean up and provide the synthetic experiments yet. However, this is the code we...

We run sinkhorn for a very large number of fixed iterations.

We run sinkhorn both for 100 and 1000 iterations and they both perform equally well.

The `pytorch_spline_conv` hard dependency will be removed in the next PyG release (targeted to land in 2-3 weeks). Sorry for the inconveniences!

The `pytorch_spline_conv` hard dependency will be removed in the next PyG release (targeted to land in 2-3 weeks). Sorry for the inconveniences!

Do you mean the smallest radius to all other nodes?