Heterogenous GAT for Graph Classification

I understand there’s an example for node classification task using HAN , but it would be great if there’s an example implementation for Graph classification task. I have some problems while implementing it - basically because the output that the model is giving is a tensor that considers the number of nodes - when i need a tensor that considers the number of graphs. Here’s an example in my repository that I can’t seem to get it working for graph classification. Any help will be highly appreciated!

See if the discussion in this thread is helpful.

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