I have an adjacency matrix with weight of edges, I donot have an edge feature. How can I include my edge weights in DGNConv instead of edge_features?
Its link is DGNConv — DGL 1.1.1 documentation
The original computation in DGNConv doesn’t include edge weights. Could you elaborate on how you want your edge weights to participate in the computation?
With edge weights, performance become improve more. That’s why I want to use. I donot have edge features just edge weights of the graph.
This topic was automatically closed 30 days after the last reply. New replies are no longer allowed.