k_hop_subgraph() not giving all subgraph edges

[yashpaneliya]

I am new to the graph neural network domain and trying my hands on basics. I was trying to generate a k-hop subgraph using k_hop_graph() method. I am not getting the desired output for it.

edge_index = torch.tensor([[0, 0, 0, 0, 1, 2, 3, 4, 5, 6, 7, 8],
                            [1, 2, 3, 4, 5, 6, 7, 8, 12, 9, 10, 11]])

subset, one_hop_edge, mapp, one_mask = k_hop_subgraph(2, 1, edge_index=edge_index)

print(one_hop_edge)

The desired output should be the edges [[0,2], [2,6]] but I'm getting only [2,6]. Can someone help me with this?

[xnuohz]

Your output is not right, it should be only [0, 2]. Node 6 cannot reach node 2 in your edges, I think torch_geometric.utils.to_undirected may be what you want.