I am confused whether to use f1 score with 'micro' average or 'macro' average for better evaluation. Given my dataset is highly imbalanced(600:100000)


1 Answer 1


To summarise this answer,

Macro calculates and F1 score for each class then averages them.

Micro calculates the recall/precision for each class, averages them then calculates the F1 score.

Micro average is preferable in a multi class situation where one class heavily outweighs the others.


Your Answer

By clicking “Post Your Answer”, you agree to our terms of service and acknowledge you have read our privacy policy.

Not the answer you're looking for? Browse other questions tagged or ask your own question.