Shape mismatch error in the GraphConv layer

Open Oxer11 opened this issue 4 years ago • 0 comments

https://github.com/DeepGraphLearning/torchdrug/blob/5bf0a505282c5e93fbf92a91ace8fe40ecc13d4c/torchdrug/layers/conv.py#L154-L174 In the message_and_aggregate() method of GraphConv class, the edge list is concatenated with self-loops while the edge input isn't. This will cause a shape mismatch error in this line.

https://github.com/DeepGraphLearning/torchdrug/blob/5bf0a505282c5e93fbf92a91ace8fe40ecc13d4c/torchdrug/layers/conv.py#L170-L171

Oct 27 '21 03:10 Oxer11