Version 0.6.9

v0.6.9 - 2025-01-06

Added

• refitting of the intercept term after fitting the rest of the model to improve the intercept value
• new options for handling missing values: "low", "high", "separate", and "gain"
• use Fischer (1958) for handling categorical values. This is the same method employed by LightGBM.
• added new parameters to control overfitting of nominal categoricals: gain_scale, min_cat_samples, cat_smooth

Changed

• enable AVX-512 by default
• modified default EBM parameters: outer_bags=16, n_jobs=-1

Fixed

• fixed memory leak in the purification function