mvpa2.algorithms.searchlight_hyperalignment.FeatureSelectionHyperalignment

class mvpa2.algorithms.searchlight_hyperalignment.FeatureSelectionHyperalignment(ref_ds=0, hyperalignment=Hyperalignment(ref_ds=0, alpha=1.0), featsel=1.0, full_matrix=True, use_same_features=False, exclude_from_model=None, dtype='float32', **kwargs)

A helper which brings feature selection and hyperalignment in a single call

It was created to facilitate its usage in any searchlight or ROI, where hyperalignment is trained and called on the same list of datasets. But unlike Hyperalignment, it doesn’t need to be separately trained. Called with a list of datasets, it returns a list of Mappers, one per each dataset, similarly to how Hyperalignment does.

Attributes

descr

Description of the object if any

Methods

__call__(datasets)
reset()

For description of parameters see SearchlightHyperalignment

Attributes

descr

Description of the object if any

Methods

__call__(datasets)
reset()