Pre processor

fyt.core.processing.pre_processor

PreProcessor

Bases: BaseEstimator, TransformerMixin

PreProcessor class that builds and executes a scikit-learn pipeline.

The pipeline performs data preprocessing based on the provided configuration.

pipeline property

The preprocessing pipeline.

Returns:

Name	Type	Description
ColumnTransformer	ColumnTransformer	The preprocessing pipeline.

random_state property writable

The random state.

fit(X, y=None)

Fit the preprocessing pipeline.

Parameters:

Name	Type	Description	Default
X	DataFrame | DataFrame	Input features.	required
y	Series | Series | ndarray | None	Target variable (optional).	None

Returns:

Name	Type	Description
PreProcessor	PreProcessor	Fitted PreProcessor instance.

handle_zero_imputation(X)

Handle zero imputation if configured.

Converts zeros to NaN only for features that are targeted by imputation components with impute_zeros=True.

Parameters:

Name	Type	Description	Default
X	DataFrame | DataFrame	Input features.	required

Returns:

Type	Description
	Features with zeros converted to NaN for specified features only.

Raises:

Type	Description
TypeError	If X is neither a polars nor a pandas DataFrame.

transform(X)

Transform the data using the fitted pipeline.

Parameters:

Name	Type	Description	Default
X	DataFrame | DataFrame	Input features to transform.	required

Returns:

Type	Description
	Transformed features.

Raises:

Type	Description
RuntimeError	If the PreProcessor has not been fitted yet.