I am going through the implementation of a graph convolutional neural network. I came across a non-binary adjacency matrix in the case of a directed graph.
The particular issue is discussed here in the following
https://github.com/tkipf/pygcn/issues/3
Can someone explain what a non-binary adjacency matrix looks like? How can an adjacency matrix have something other than 0 or 1?
In some uses, the value of the adjacency matrix is the number of parallel edges connecting the nodes.