two_step_habitat

Note

This page is a reference documentation. It only explains the function signature, and not how to use it. Please refer to the Habitat Guide and Python API guide (v2.0) for usage.

two_step_habitat(*, modalities: Sequence[str], n_supervoxels: int = 50, n_habitats: int | str = 'auto', habitat_features: Sequence[str | Spec | Mapping[str, object]] | None = None, random_seed: int | None = None, supervoxel_algorithm: str = 'kmeans', habitat_fitter_algorithm: str = 'kmeans', roi: str = 'tumor') → Study[source]

Declare a classical two-step habitat study.

Parameters:

  • modalities – Modality names for the raw voxel extractor.

  • n_supervoxels – Number of supervoxels per subject.

  • n_habitats – Fixed habitat count or "auto" with elbow search.

  • habitat_features – Optional habitat feature families ("msi", etc.).

  • random_seed – Seed for every seedable component.

  • supervoxel_algorithm – Registered supervoxelizer name.

  • habitat_fitter_algorithm – Registered cohort fitter name.

  • roi – ROI keyword for voxel extraction.

Returns:

A Study ready for Study.fit().

See also

habit.recipes.Study

Sklearn-style fit / fit_predict / predict entry.

habit.spec.HabitatSpec

Frozen analysis declaration the factory builds.

habit.recipes.one_step_habitat

Per-subject habitat definition.

habit.recipes.direct_pooling_habitat

Cohort clustering on voxel features.

Examples using habit.recipes.two_step_habitat

Quickstart: Python API

Quickstart: Python API