Stuck on an issue?

Lightrun Answers was designed to reduce the constant googling that comes with debugging 3rd party libraries. It collects links to all the places you might be looking at while hunting down a tough bug.

And, if you’re still stuck at the end, we’re happy to hop on a call to see how we can help out.

AssertionError assert row.max().item() < sparse_sizes[0] after updating

See original GitHub issue

Hi,

I was using the function torch_scatter.coalesce before with no problems. I updated today to the lasted version of pytorch geometric available via conda and now I get this error message: assert row.max().item() < sparse_sizes[0] AssertionError I have checked the torch_sparse.coalesce history and it seems this part of the code has been modified recently https://github.com/rusty1s/pytorch_sparse/blob/8c25ddefdc9f6e797aafff5782d1840025f80fa2/torch_sparse/storage.py#L21. Is there a bug in the latest version installed via conda ? Or it must be something I am doing wrong? Thanks

Issue Analytics

State:
Created 2 years ago
Comments:5 (2 by maintainers)

Top GitHub Comments

1reaction

mariolinovICcommented, Jul 28, 2021

I dont have problems with pool_edge, but I have a similar function to it. And in my case I had problems due to the indexing I was using for the nodes did not meet the assert condition. Altough I think that for a standard use it should not cause issues as you pointed. So no worries. Thank you very much for the help.

0reactions

rusty1scommented, Jul 28, 2021

That is correct (although the output will always be correct anyway since cluster.size(0) > cluster.max(). Do you wanna fix it anyway?