Hi everyone. I was reading the book “Graph Representation Learning” and the author mentions about possible node features such as node degree, node centrality, and clustering coefficient.
Are any of these possible node features already available in DGL somehow?
DGL contains DGLGraph.in_degrees and DGLGraph.out_degrees for querying node degrees. For centrality and clustering coefficient APIs, you may refer to cugraph or networkx (DGL provides graph conversion from and to them).
This topic was automatically closed 30 days after the last reply. New replies are no longer allowed.