API reference

coming soon

Programmatic entry points behind the challenge library commands.

epibench.library

all_challenges() -> dict[str, dict]

Return {challenge_id: definition} for every challenge JSON in the bundled library, sorted by id.

load_challenge(challenge_id: str) -> dict

Load one challenge definition by id; raises if the id is not in the library.

is_published(definition: dict) -> bool

True when a challenge has a real Zenodo DOI (i.e. data available to download).

epibench.fetch

fetch(challenge_id: str, output_path: str | None = None) -> None

Download a challenge's data files from Zenodo into <output_path>/<challenge_id>/ (defaults to the current directory), unzipping archives and keeping a copy of the challenge definition.