Datasets

gene_calling_benchmark.load_dataset(name)

Fetch (or reuse cached) dataset and return resolved file paths.

Return type:

LoadedDataset

Parameters:

name (str)

gene_calling_benchmark.list_datasets()

Names of all datasets in the registry, in declaration order.

Return type:

list[str]

gene_calling_benchmark.get_dataset_info(name)

Return the DatasetSpec for name without downloading.

Return type:

DatasetSpec

Parameters:

name (str)

class gene_calling_benchmark.LoadedDataset(name, fasta, annotation, labels, spec)

Resolved on-disk paths for a benchmark dataset.

Parameters:

• name (str)

• fasta (Path)

• annotation (Path)

• labels (Path | None)

• spec (DatasetSpec)

as_tuple()

Convenience: (fasta, annotation) for tools that take exactly those two.

Return type:

tuple[Path, Path]