pretab.transformers.CyclicalTimeTransformer

class pretab.transformers.CyclicalTimeTransformer(period)

Bases: BasePreTabTransformer

Encode a cyclical time variable using sine and cosine components.

Maps a periodic integer feature (such as hour of day or day of week) onto two continuous features so that the cyclic boundary is continuous.

Parameters:

period (int) – The full cycle length (e.g., 24 for hours, 7 for weekdays).

Notes

For a value \(x\) with period \(p\), the encoding is

\[\left(\sin\!\left(\frac{2\pi x}{p}\right),\; \cos\!\left(\frac{2\pi x}{p}\right)\right).\]

Each input feature therefore expands into two output columns.

This is a standalone time-series utility. Although it preserves the row count, it takes a required per-feature period and constrains inputs to [0, period], so it is not wired into Preprocessor (which applies one method uniformly across columns). Apply it directly to the relevant cyclical column instead.

Examples

>>> import numpy as np
>>> from pretab.transformers import CyclicalTimeTransformer
>>> X = np.array([[0], [6], [12], [18]])
>>> transformer = CyclicalTimeTransformer(period=24)
>>> transformer.fit_transform(X).shape
(4, 2)

__init__(period)

Methods

__init__(period)
fit(X[, y])
fit_transform(X[, y])	Fit to data, then transform it.
get_feature_names_out([input_features])	Return output feature names of the form {feature}_{suffix}{j}.
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)

Attributes

total_output_dim_	Total number of output columns produced across all input features.
n_features_in_
adaptive