RGCN-pytorch
RGCN-pytorch copied to clipboard
berlincho
Pytorch implementation of Relational GCN for node classification
make it compatible with PyTorch 2 and updating README.md accordingly. Also excluded downloaded files from repositories in .gitignore
Hi, in this repo, all four datasets under `\data` are quite small. Can R-GCN scale to very large datasets (e.g. containing million edges) without memory issues? Thanks!
does r-gcn works on entity alignment (existing R-GCN seems only focus on classification and link prediction)? If yes, May I know do you have any working code on this?
https://github.com/berlincho/RGCN-pytorch/blob/14f9c633b1637fe41c91e757887a43017ff53db2/data_utils.py#L230-L232 you are assuming these 3 files are on the local, where do they come from?
After training the R-GCN on a dataset, how do I get the entity embeddings?
