Question about Sampling Multiple Edge Types Simultaneously in Heterogeneous Graphs using PyG #7551
Unanswered
ccoverflow
asked this question in
Q&A
Replies: 1 comment 7 replies
-
Currently |
Beta Was this translation helpful? Give feedback.
7 replies
Sign up for free
to join this conversation on GitHub.
Already have an account?
Sign in to comment
Uh oh!
There was an error while loading. Please reload this page.
Uh oh!
There was an error while loading. Please reload this page.
-
Hi, I have a heterogeneous graph data that has been converted to the HeteroData class. This heterogeneous graph data has more than 25 types of edges, each with a timestamp. I need to sample multiple types of edges simultaneously when loading (for example, for a type 1 edge, select other edges within its designated hops (like 2 hops), which have a common parent node and whose occurrence time satisfies the chronological order, to be the neighbors of edge 1, and set to sample n neighbors of edge 1). However, the 'edge_label_index' parameter in 'torch_geometric.loader.link_neighbor_loader' can only accept one type of edge. I have also tried implemented loaders such as HGTLoader, but none of them can achieve the effect I want. I'm wondering if there are similar class methods in PyG that can achieve this goal?
Beta Was this translation helpful? Give feedback.
All reactions