PowerTransformer#

class pyts.preprocessing.PowerTransformer(method: Literal['yeo-johnson', 'box-cox'] = 'yeo-johnson', standardize: bool = True)[source]#

Apply a power transform sample-wise to make data more Gaussian-like.

Power transforms are a family of parametric, monotonic transformations that are applied to make data more Gaussian-like. This is useful for modeling issues related to heteroscedasticity (non-constant variance), or other situations where normality is desired.

Currently, PowerTransformer supports the Box-Cox transform and the Yeo-Johnson transform. The optimal parameter for stabilizing variance and minimizing skewness is estimated through maximum likelihood.

Box-Cox requires input data to be strictly positive, while Yeo-Johnson supports both positive or negative data.

By default, zero-mean, unit-variance normalization is applied to the transformed data.

Parameters:
method‘yeo-johnson’ or ‘box-cox’ (default = ‘yeo-johnson’)

The power transform method. Available methods are:

  • ‘yeo-johnson’ [1], works with positive and negative values

  • ‘box-cox’ [2], only works with strictly positive values

standardizeboolean (default = True)

Set to True to apply zero-mean, unit-variance normalization to the transformed output.

Notes

NaNs are treated as missing values: disregarded in fit, and maintained in transform.

References

[1]

I.K. Yeo and R.A. Johnson, “A new family of power transformations to improve normality or symmetry.” Biometrika, 87(4), pp.954-959, (2000).

[2]

G.E.P. Box and D.R. Cox, “An Analysis of Transformations”, Journal of the Royal Statistical Society B, 26, 211-252 (1964).

Examples

>>> import numpy as np
>>> from pyts.preprocessing import PowerTransformer
>>> X = [[1, 3, 4], [2, 2, 5]]
>>> pt = PowerTransformer()
>>> print(pt.transform(X))
[[-1.316...  0.209...  1.106...]
 [-0.707... -0.707...  1.414...]]

Methods

__init__([method, standardize])

fit([X, y])

Pass.

fit_transform(X[, y])

Fit to data, then transform it.

get_metadata_routing()

Get metadata routing of this object.

get_params([deep])

Get parameters for this estimator.

set_params(**params)

Set the parameters of this estimator.

transform(X)

Transform the data.
__init__(method: Literal['yeo-johnson', 'box-cox'] = 'yeo-johnson', standardize: bool = True) → None[source]#
fit(X: ArrayLike | None = None, y: ArrayLike | None = None) → Self[source]#

Pass.

Parameters:
X

Ignored

y

Ignored

Returns:
selfobject
transform(X: ArrayLike) → NDArray[float64][source]#

Transform the data.

Parameters:
Xarray-like, shape = (n_samples, n_timestamps)

Data to transform.

Returns:
X_newarray-like, shape = (n_samples, n_timestamps)

Transformed data.

Examples using pyts.preprocessing.PowerTransformer#

Transformers

Transformers