Available in: GBM, DRF, XGBoost
calibrate_model option allows you to specify Platt scaling in GBM and DRF to calculate calibrated class probabilities. Platt scaling transforms the output of a classification model into a probability distribution over classes. It works by fitting a logistic regression model to a classifier’s scores. Platt scaling will generally not affect the ranking of observations. Logloss, however, will generally improve with Platt scaling.
calibrate_model option is disabled by default. When enabled, the calibrated probabilities will be appended to the frame with the original prediction.
Note that when this option is enabled, then you must also specify the calibration dataframe (specified with calibration_frame) that will be used for Platt scaling. A best practice is to split the original dataset into training and calibration sets.
Refer to the following for more information about Platt scaling: