def CatEncoderWorker.fit(self, inputs, fx_choice=REG.AIG, **params) [source]
Analyze and learn categorical encoding.
Executes the predictive gates, missing handling, and
min-bin-share merging for each categorical feature.
Results are stored as encoding maps ready for
transform.
Parameters
inputs : CatEncoderInput
Input configuration including train_set and
categorical feature lists.
categorical feature lists.
fx_choice : str
The categorical encoding strategy. Defaults to REG.AIG.
**params : dict
Runtime parameters passed to the internal fitter
(gini_cutoff, min_bin_share).
(gini_cutoff, min_bin_share).
Returns
CatEncoder
The fitted encoder instance.