Inductive or transductive DistDGL GraphSAGE?

Just want to confirm something:

The distributed GraphSAGE code in graphsage/experimental is doing inductive learning, correct? (i.e. during training only the training nodes are seen)

Thank you.

It’s transductive. All nodes’ features are known in the training process, but only training nodes have labels

Thanks for the info!

The GraphSage model for distributed training is inductive.

Does this mean that the single host GraphSAGE (and GCN, if I may ask as well) are transductive while the distributed one is inductive?

The single-machine version of GraphSAGE allows both transductive learning and inductive learning. GCN was originally developed for transductive learning but can be used for inductive learning as well.

This topic was automatically closed 30 days after the last reply. New replies are no longer allowed.