shen-lab / gda-specreg Goto Github PK

Dear author,
I would like to express my appreciation for your outstanding work. I have a question regarding the experimental evaluation of node classification in the ACM and DBLP datasets. I noticed that a validation set does not appear to have been used, and instead, the best test performance of ACM and DBLP was directly selected. I was wondering if this approach might have some potential limitations or if there is a specific reason for not using a validation set. I would greatly appreciate your clarification on this matter. Looking forward to your reply.
Thank you.