search

RUCAIBox / NCL

[WWW'22] Official PyTorch implementation for "Improving Graph Collaborative Filtering with Neighborhood-enriched Contrastive Learning".
117 stars 20 forks source link

ProtoNCE_Loss #34

Closed Zhuohan007 closed 8 months ago

Zhuohan007 commented 1 year ago

Hello！When calculating ProtoNCE_loss, why do you use the embedding of the 0th layer of the node instead of the embedding after the k-layer is propagated? Please reply if you see it.

linzihan-backforward commented 1 year ago

Hi~ You can reference this issue to get the answer.