Is there a good way to insert edges in a graph instance?

[bryanzhou008]

I am puzzled as I couldn't find any online answers explaining this. Which leads me to wonder if this data type is not meant to be operated on in this manner?

[rusty1s]

You can insert edged by concatenating them to your edge_index:

new_edges = torch.tensor([[0, 1, 2, 3], [1, 0, 3, 2]])
edge_index = torch.cat([edge_index, new_edges], dim=1)