Random Projections

Generate multiple projected views from a single feature matrix.

Utilities for generating multiple random-projection views from a single matrix.

class polyview.augmentation.random_projections.RandomProjectionViews(*args: Any, **kwargs: Any)

Bases: BaseEstimator

Generate multiple projected views from a single matrix.

This transformer is pipeline-friendly: it accepts a single 2-D matrix and outputs a MultiViewDataset containing n_views independent random projections of the input.

fit(X, y=None) → RandomProjectionViews

fit_transform(X, y=None) → MultiViewDataset

transform(X) → MultiViewDataset

polyview.augmentation.random_projections.random_projection(X: numpy.ndarray | Sequence[Sequence[float]], n_views: int = 2, n_components: int | Sequence[int | None] | None = None, method: Literal['gaussian', 'sparse'] = 'gaussian', random_state: int | None = None, labels=None, view_names: Sequence[str] | None = None) → MultiViewDataset

Create multiple projected views from a single 2-D matrix.

This helper is equivalent to instantiating RandomProjectionViews and calling fit_transform.