gtda.mapper.method_to_transform(cls, method_name)[source]

Wrap a class to add a transform method as an alias to an existing method.

An example of use is for classes possessing a score method such as kernel density estimators and anomaly/novelty detection estimators, allow for these estimators are to be used as steps in a pipeline.

Note that 1D array outputs are reshaped into 2D column vectors before being returned by the new transform.

  • cls (object) – Class to be wrapped. If method_name is not one of its methods, transform always returns None.

  • method_name (str) – Name of the method in cls to which transform will be an alias. The fist argument of this method (after self) becomes the X input for transform.


wrapped_cls – New class inheriting from sklearn.base.TransformerMixin, so that both transform and fit_transform are available. Its name is the name of cls prepended with 'Extended'.

Return type



>>> import numpy as np
>>> from sklearn.neighbors import KernelDensity
>>> from gtda.mapper import method_to_transform
>>> X = np.random.random((100, 2))
>>> kde = KernelDensity()

Extend KernelDensity to give it a transform method as an alias of score_samples (up to output shape). The new class is instantiated with the same parameters as the original one.

>>> ExtendedKDE = method_to_transform(KernelDensity, 'score_samples')
>>> extended_kde = ExtendedKDE()
>>> Xt =
>>> print(Xt.shape)
>>> Xt_extended = extended_kde.fit_transform(X)
>>> print(Xt_extended.shape)
(100, 1)
>>> np.array_equal(Xt, Xt_extended.flatten())