Closed mohitbadwal closed 7 years ago
This is intentional, because AdaBoostClassifier implements some pseudo-probabilistic classification algorithm (the predicted class and the predicted probability distribution do not "match").
I would suggest using some other boosting algorithm, such as XGBoost or LightGBM. They both provide excellent Scikit-Learn interoperability.
Ensemble has AdaBoostRegressor but it does not have AdaBoostClassifier , what should be done?