Open veralauee opened 5 years ago
This used the learned metric, fc wasn't reported in the main paper even though it would probably do better. To be clear, this is the Maryland Polyvore's original splits with the random negatives, and not our resampled version.
I am confused about the exact setting of "args" to achieve the best performance (fitb acc= 86.1, comp auc=0.98) in your paper. Would you please tell me how to set these arguments? For example, --use_fc, --learned_metric, etc.