gluonts.dataset.arrow.file module

class gluonts.dataset.arrow.file.ArrowFile(path: pathlib.Path, _start: int = 0, _take: Union[int, NoneType] = None)

Bases: gluonts.dataset.arrow.file.File

property batch_offsets

decoder: gluonts.dataset.arrow.dec.ArrowDecoder

iter_batches()

location_for(idx)

metadata() → Dict[str, str]

path: pathlib.Path

reader: pyarrow.ipc.RecordBatchFileReader

property schema

class gluonts.dataset.arrow.file.ArrowStreamFile(path: pathlib.Path, _start: int = 0, _take: Union[int, NoneType] = None)

Bases: gluonts.dataset.arrow.file.File

metadata() → Dict[str, str]

path: pathlib.Path

class gluonts.dataset.arrow.file.File

Bases: object

SUFFIXES = {'.arrow', '.feather', '.parquet'}

static infer(path: pathlib.Path) → Union[gluonts.dataset.arrow.file.ArrowFile, gluonts.dataset.arrow.file.ArrowStreamFile, gluonts.dataset.arrow.file.ParquetFile]

Return ArrowFile, ArrowStreamFile or ParquetFile by inspecting provided path.

Arrow’s random-access format starts with ARROW1, so we peek the provided file for it.

abstract metadata() → Dict[str, str]

class gluonts.dataset.arrow.file.ParquetFile(path: pathlib.Path, _start: int = 0, _take: Union[int, NoneType] = None, _row_group_sizes: List[int] = <factory>)

Bases: gluonts.dataset.arrow.file.File

location_for(idx)

metadata() → Dict[str, str]

path: pathlib.Path

reader: pyarrow.parquet.core.ParquetFile