ChandlerBang/Pro-GNN

About pubmed dataset

Closed this issue · 1 comments

songshuhan commented

Hi，Thank you so much for sharing the code，However, when I run large data sets such as PubMed, there will be a problem of insufficient CUDA memory,How did you solve it？

ChandlerBang commented

Hi,

Pro-GNN needs to learn a dense adjacency matrix which requires large GPU memory just as attacking the adjacency matrix. You may refer to #7.