Some problems for TANE

Open leepp199 opened this issue 1 year ago • 0 comments

I followed the code you provided and performed the replication experiments on miniImage dataset, but the results I got have higher values of acc and auroc for ATT scheme than ATT-G, SEMAN-G scheme, I would like to ask you if you didn't provide the complete hyper-parameters for training? Or is it some other reason? 1shot_att 1shot_attg 1shot_semang 5shot_att 5shot_attg 5shot_semang Paper's result for FSOR

Oct 18 '24 06:10 leepp199