TSCPolynomialFeatures#

class datafold.dynfold.TSCPolynomialFeatures(degree=2, *, interaction_only=False, include_bias=False, include_first_order=False)[source]#

Bases: PolynomialFeatures, TSCTransformerMixin

Compute polynomial features from data.

This is a subclass of PolynomialFeatures from scikit-learn to generalize the input and output of pandas.DataFrames and TSCDataFrame.

This class adds the parameter include_first_order to choose whether to include the identity states. For all other parameters please visit the super class documentation of PolynomialFeatures.

Attributes Summary

powers_

Exponent for each of the inputs in the output.

Methods Summary

`fit`(X[, y])	Compute number of output features.
`partial_fit`(X[, y])
`transform`(X)	Transform data to polynomial features.

Attributes Documentation

powers_#

Methods Documentation

fit(X, y=None, **fit_params)[source]#

Compute number of output features.

Parameters:

X (TSCDataFrame, pandas.DataFrame, numpy.ndarray) – Data of shape (n_samples, n_features).
y (None) – ignored
**fit_params (Dict[str, object]) – None

Returns:

self

Return type:

TSCPolynomialFeatures

partial_fit(X, y=None, **fit_params)[source]#

transform(X)[source]#

Transform data to polynomial features.

Parameters:: X (TSCDataFrame, pandas.DataFrame, numpy.ndarray) – The data of shape (n_samples, n_features) to transform.
Returns:: Transformed data of shape (n_samples, n_polynomials) and with same type as X.
Return type:: TSCDataFrame, pandas.DataFrame, numpy.ndarray