Mean binarizer

fyt.core.processing.mean_binarizer

MeanBinarizer

Bases: BaseEstimator, TransformerMixin

Discretizes numerical features based on their mean values.

This transformer binarizes each feature by comparing values to the mean computed during fitting. Values above the mean are encoded as 1, values at or below the mean as 0. The per-feature means are stored on the fitted instance as mean_ (ndarray of shape (n_features,)).

Examples:

>>> import numpy as np
>>> from fyt.core.processing.mean_binarizer import MeanBinarizer
>>> X = np.array([[1, 2], [3, 4], [5, 6]])
>>> binarizer = MeanBinarizer()
>>> binarizer.fit(X)
>>> binarizer.transform(X)
array([[0, 0],
       [0, 0],
       [1, 1]])

fit(X, y=None)

Compute the mean for each feature.

Parameters:

Name	Type	Description	Default
X	DataFrame | ndarray	array-like or DataFrame of shape (n_samples, n_features) Training data.	required
y	object	Ignored. Present for sklearn compatibility.	None

Returns:

Name	Type	Description
self	'MeanBinarizer'	Returns the instance itself.

set_output(*, transform=None)

Set output configuration for compatibility with sklearn pipelines.

Parameters:

Name	Type	Description	Default
transform	str | None	Output format for transform method. Not used in this wrapper.	None

Returns:

Name	Type	Description
MeanBinarizer	'MeanBinarizer'	The transformer itself.

transform(X)

Transform features to binary values based on the fitted mean.

Parameters:

Name	Type	Description	Default
X	DataFrame | ndarray	array-like or DataFrame of shape (n_samples, n_features) Data to transform.	required

Returns:

Name	Type	Description
X_transformed		ndarray or DataFrame of shape (n_samples, n_features) Binarized data. Values > mean are 1, values <= mean are 0. Returns pl.DataFrame if input was pl.DataFrame, otherwise ndarray.