DirectoryDataSource

Note

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

class DirectoryDataSource(root: str | Path, *, modalities: Sequence[str], roi: str, images_folder: str = 'images', masks_folder: str = 'masks', name: str | None = None, metadata: Dict[str, Any] | None = None)[source]

Bases: object

Build a Cohort from HABIT’s conventional directory layout.

Parameters:

  • root – Directory root holding the images and masks subdirectories.

  • modalities – Modality keys to include, in analysis order. A subject missing any requested modality is skipped with a warning, mirroring the v0.1 scan behaviour.

  • roi – Mask key identifying the region of interest.

  • images_folder – Name of the images subdirectory under root.

  • masks_folder – Name of the masks subdirectory under root.

  • name – Human-readable cohort name used in reports.

  • metadata – Optional cohort-level attributes (centre, scanner, study).

__init__(root: str | Path, *, modalities: Sequence[str], roi: str, images_folder: str = 'images', masks_folder: str = 'masks', name: str | None = None, metadata: Dict[str, Any] | None = None) → None[source]
load() → Cohort[source]

Build the cohort described by this source.

Returns:

A cohort with a defined, reproducible subject order (sorted subject ids), whose images and masks are lazy file references.

Raises:

DataFormatError – If the convention folders are missing.