Graph-Classification-with-GCN
Graph-Classification-with-GCN copied to clipboard
Featureless flag bug
Hello, When I only edit the "featureless" flag to be true and leave everything else the same, suddenly the training accuracy becomes 1.0 (on the default protein dataset). Help!
Thank you for your help, i'm stil looking for the bug. I noticed that the accuracy drops a little if you use less training iteration, it reaches 0.9 too. While i search, if you find what's going on i'd really appreciate if you would point out to me!