I have an adjaceny tensor
tensor([[0., 1., 1., 1.],
[1., 0., 1., 1.],
[1., 1., 0., 1.],
[1., 1., 1., 0.]])
How can I convert it to the edge_index
form used in pytorch geometric? Introduction by Example — pytorch_geometric 2.0.4 documentation The actual adjacency matrix is far larger so it’s not something I want to create manually.