Train Cora with the sparse version

PetarV- / GAT

Graph Attention Networks (https://arxiv.org/abs/1710.10903)

https://petar-v.com/GAT/

MIT License

3.18k stars 642 forks source link

Closed gcucurull closed 6 years ago

gcucurull commented 6 years ago

This PR defines a sparse GAT using the sparse head (with some minor modifications) and trains it on Cora.