gluonts.dataset

gluonts.dataset.DataEntry

gluonts.dataset.DataBatch

class gluonts.dataset.Dataset(*args, **kwds)

class gluonts.dataset.DatasetCollection(*args, **kwds)

Flattened access to a collection of datasets.

datasets: List[Dataset]

interleave: bool = False

iter_interleaved()

iter_sequential()

class gluonts.dataset.DatasetWriter

write_to_file(dataset: Dataset, path: pathlib.Path) → None

write_to_folder(dataset: Dataset, folder: pathlib.Path, name: Optional[str] = None) → None

Dataset Writers

class gluonts.dataset.DatasetWriter

write_to_file(dataset: Dataset, path: pathlib.Path) → None

write_to_folder(dataset: Dataset, folder: pathlib.Path, name: Optional[str] = None) → None

Arrow

class gluonts.dataset.arrow.ArrowWriter(stream: bool = False, suffix: str = '.arrow', flatten_arrays: bool = True, metadata: Union[dict, NoneType] = None)

class gluonts.dataset.arrow.ParquetWriter(suffix: str = '.parquet', flatten_arrays: bool = True, metadata: Union[dict, NoneType] = None)

JsonLines

class gluonts.dataset.jsonl.JsonLinesWriter(use_gzip: bool = True, suffix: str = '.json', compresslevel: int = 4)