Can FastGAE be used in weighted graph?

I hope to use graph variational autoencoders in weighted biological networks with hundreds of thousands of nodes and millions of edges. FastGAE can help handle such a large network, but we noticed $A_{i,j} \in$ { 0, 1 } in Proposition 4.

Could you please tell me whether FastGAE can be used for weighted networks?


The issue points to Proposition 4 in the FastGAE article and provides an image, but names no repository file or test. Start by checking the implementation and derivation around the binary adjacency assumption; done means clarifying whether weighted biological networks are supported or identifying the required scope.

Written by the indexing model from the issue text.