How to embed non-graph features

Now I am working on a regression problem. However, each sample has both graph features and non-graph features (such as flow rate, ratio), how to embed these non-graph features into a graph neural network?
Thank you very much!

You could first compute the representation of a graph using a GNN + graph pooling (SumPooling, MaxPooling, etc.) and then concatenate them with the non-graph features, then pass them into an MLP to get the prediction.

OH, Thank you bro very much.

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