Open Mobzhang opened 3 years ago
How big is your graph? I guess that memory will overflow because of the creation of the neg_adj_mask
, which is a dense bool tensor of shape [num_nodes, num_nodes]
. You may want to remove the creation of this matrix and replace it with negative_sampling
to see if that fixes your issues.
thanks for your reply. my graph has19K nodes 144K edges. I'll try it based on your advice.
When I use train_test_split_edges method to split a large graph, memory will overflow. What should I do?