AUCPR on Multi-Class dataset

#1

Hello,

I’m training a multi-class model with XGBoost Classifier and I’ve set the eval_metric to ‘aucpr’. In a post from 2020 on this forum, I see that ‘aucpr’ is not intended for multi-class training but I’m wondering if that’s been updated? If so, does it now give macro-weighted AUCPR scores? Thank you!

#2

@hcho3 , Hey I hope you’re doing well. I was wondering if you had any input on this. Thank you!