datumaro.plugins.data_formats.arrow#

class datumaro.plugins.data_formats.arrow.ArrowBase(root_path: str, *, file_paths: List[str], ctx: ImportContext | None = None)[source]#

Bases: DatasetBase

infos() → Dict[str, Any][source]#: Returns meta-info of dataset.

categories() → Dict[AnnotationType, Categories][source]#: Returns metainfo about dataset labels.

get(item_id: str, subset: str | None = None) → DatasetItem | None[source]#: Provides random access to dataset items.

property lookup: Dict[str, Dict[str, int]]#

subsets() → Dict[str, IDataset][source]#: Enumerates subsets in the dataset. Each subset can be a dataset itself.

get_subset(name: str) → IDataset[source]#

class datumaro.plugins.data_formats.arrow.ArrowExporter(extractor: IDataset, save_dir: str, *, save_media: bool | None = None, image_ext: str | Callable[[str], bytes] | None = None, default_image_ext: str | None = None, save_dataset_meta: bool = False, ctx: ExportContext | None = None, num_workers: int = 0, max_shard_size: int | None = 1000, num_shards: int | None = None, prefix: str = 'datum', **kwargs)[source]#

Bases: Exporter

AVAILABLE_IMAGE_EXTS = ('AS-IS', 'PNG', 'TIFF', 'JPEG/95', 'JPEG/75', 'NONE')#

DEFAULT_IMAGE_EXT = 'AS-IS'#

classmethod build_cmdline_parser(**kwargs)[source]#

classmethod patch(dataset, patch, save_dir, **kwargs)[source]#

class datumaro.plugins.data_formats.arrow.ArrowImporter[source]#

Bases: Importer

classmethod detect(context: FormatDetectionContext) → FormatDetectionConfidence | None[source]#

classmethod find_sources(path: str) → List[Dict][source]#

classmethod find_sources_with_params(path: str, **extra_params) → List[Dict][source]#

classmethod get_file_extensions() → List[str][source]#

Modules

`datumaro.plugins.data_formats.arrow.base`
`datumaro.plugins.data_formats.arrow.exporter`
`datumaro.plugins.data_formats.arrow.format`
`datumaro.plugins.data_formats.arrow.importer`
`datumaro.plugins.data_formats.arrow.mapper`