Can you specify how AUC is being calculated? In particular, I'm wondering if the predictions need to specifically align with the 0-1 range. This is not the case for the sklearn implementation where only the ordering of the predictions matters, so, for example: roc_auc_score(y_true, y_score) = roc_auc_score(y_true, y_score*35)
Posted by: ryanv777 @ May 18, 2022, 4:22 a.m.Hi,
The predictions should be in the 0-1 range.
You can find more information on how the metric roc_auc_score is calculated on the Sklearn page.
(https://scikit-learn.org/stable/modules/generated/sklearn.metrics.roc_auc_score.html)
MAFAT Challenge Team
Posted by: MAFAT_Challenge @ May 18, 2022, 11:59 a.m.