ertk.dataset.load_multiple

ertk.dataset.load_multiple(corpus_files: Iterable[str | PathLike], features: str | None = None, subsets: str | Mapping[str, str] = 'default', label: str | Mapping[str, str] = 'label', **read_kwargs) → CombinedDataset

Load one or more datasets with the given features.

Parameters:

corpus_files: iterable

The corpus description YAML files to load.

features: str

A common set of features to load. This will be found in the features directory corresponding to each corpus.

subsets: str or dict

A subset name common to all datasets (e.g. “all”, “default”) or a mapping from dataset name to subset name.

**read_kwargs:

Other args to pass to feature loading.