Solved – Probability for class in xgboost

Question

I'm using xgboost for a problem where the outcome is binary but I am only interested in the correct probability of a sample to be in class 1.

My current approach is to use the XGBClassifier in Python with objective binary:logistic, use predict_proba method and take that output as a probability for class 1.

I that methodology correct in the sense that I get actual probability values for a sample to be in class 1?

Answer 1

predict_proba yields estimated probabilities that a sample is in class 1.

Note that the speaker in the comment is the author of xgboost, so this is the definitive answer on the subject.