Target Based Features

class target_based_features.TargetBasedEncoder(method: str = 'target_mean', target_col: str = '', group_col: str = '', smoothing: int | None = None, n_splits: int = 5)

A transformer that provides various methods for creating target-based features, including target mean encoding, smoothed target mean encoding, count encoding, cross-validated target encoding, and Weight of Evidence (WoE).

fit(X: DataFrame, y: Series | None = None) → TargetBasedEncoder

Fit method for compatibility with the scikit-learn API.

Parameters:

• X (pd.DataFrame) – Input DataFrame.

• y (pd.Series, optional) – Target variable (ignored).

Returns:

Returns self.

Return type:

TargetBasedEncoder

transform(X: DataFrame) → DataFrame

Applies the selected encoding method to the DataFrame.

Parameters:

X (pd.DataFrame) – Input DataFrame.

Returns:

DataFrame with the target-based feature encoding applied.

Return type:

pd.DataFrame