ashrefm / multi-label-soft-f1

Train a multi-label image classifier with macro soft-F1 loss in TensorFlow 2.0

Geek Repo:Geek Repo

Github PK Tool:Github PK Tool

About f1 score(micro or macro)

logicmixtape opened this issue · comments

Thank you for your detailed and thoughtful post!
I have one question.
it appears that there is an imbalance in the number of labals.
Shouldn't you use micro instead of macro at this point??