pretab.transformers.CustomBinTransformer

class pretab.transformers.CustomBinTransformer(output_dim=UNSET)[source]

Bases: AliasResolverMixin, TransformerMixin, BaseEstimator

Custom binning transformer for one-dimensional numerical features.

This transformer bins continuous values into discrete intervals, using either a fixed number of equal-width bins or a user-provided array of bin edges. It is compatible with scikit-learn pipelines.

Parameters:

output_dim (int or array-like) – If int, defines the number of equal-width bins. If array-like, defines the bin edges to use directly. Note that output_dim here is the number of bins, not the number of output columns: this transformer always emits a single ordinal column of integer bin indices. The bin count only becomes an output width after a subsequent one-hot encoding.

Variables:

n_features_in (int) – The number of input features seen during fit (expected to be 1).
total_output_dim (int) – Total number of output columns (fitted). Always 1 because the output is a single ordinal column.

Notes

This transformer operates on a single feature of shape (n_samples, 1). When output_dim is an integer, equal-width bin edges are computed from the data range; when it is an array-like, the provided edges are used directly. The output contains integer bin indices in a single column, so its width is 1 regardless of output_dim – this is a documented exception to the exact-width contract that the fixed-basis families follow.

The input must be numeric: binning is performed with pandas.cut(), so string / categorical data cannot be processed and raises a PretabDataError. Encode such columns with a categorical method (e.g. "int" or "one-hot") before binning.

Examples

>>> import numpy as np
>>> from pretab.transformers import CustomBinTransformer
>>> X = np.linspace(0, 1, 10).reshape(-1, 1)
>>> transformer = CustomBinTransformer(output_dim=4)
>>> transformer.fit_transform(X).shape
(10, 1)

__init__(output_dim=UNSET)[source]

Methods

`__init__`([output_dim])
`fit`(X[, y])	Fit the transformer on the data.
`fit_transform`(X[, y])	Fit to data, then transform it.
`get_feature_names_out`([input_features])	Return the names of the transformed features.
`get_metadata_routing`()	Get metadata routing of this object.
`get_params`([deep])	Get parameters for this estimator.
`set_output`(*[, transform])	Set output container.
`set_params`(**params)	Set the parameters of this estimator.
`transform`(X)	Transform the data using the specified binning strategy.

fit(X, y=None)[source]

Fit the transformer on the data.

Parameters:

X (array-like of shape (n_samples, 1)) – Input data.
y (Ignored) – Not used, present here for API consistency by convention.

Returns:

self (object) – Fitted transformer.

get_feature_names_out(input_features=None)[source]

Return the names of the transformed features.

Parameters:: input_features (list of str) – The names of the input features.
Returns:: input_features (ndarray of shape (n_features,)) – The names of the output features after transformation.

transform(X)[source]

Transform the data using the specified binning strategy.

Parameters:: X (array-like of shape (n_samples, 1)) – Input data to transform.
Returns:: X_binned (ndarray of shape (n_samples, 1)) – Binned data with integer bin indices.