allow non-accuracy scoring for TAO

csinva / imodels

Interpretable ML package 🔍 for concise, transparent, and accurate predictive modeling (sklearn-compatible).

https://csinva.io/imodels

MIT License

1.39k stars 122 forks source link

Closed keyan3 closed 2 years ago

keyan3 commented 2 years ago

allow other sklearn metrics, such as 'roc_auc_score' or 'average_precision' to be used for scoring TAO updates