roc_auc score GridSearch

Question

I am experimenting with xgboost.

I ran GridSearchCV with score='roc_auc' on xgboost. The best classificator scored ~0.935 (this is what I read from GS output). But now when I run best classificator on the same data:

roc_auc_score(Y, clf_best_xgb.predict(X))

it gives me score ~0.878

Could you tell me how the score is evaluated in both cases? Thanks

Answer 1

Try using predict_proba instead of predict as below. It should give you the same number.

roc_auc_score(Y, clf_best_xgb.predict_proba(X)[:,1])

When we compute AUC, most of time people will use the probability instead of the actual classs.