Attention

June 2024 Status Update: Removing DataPipes and DataLoader V2

We are re-focusing the torchdata repo to be an iterative enhancement of torch.utils.data.DataLoader. We do not plan on continuing development or maintaining the [DataPipes] and [DataLoaderV2] solutions, and they will be removed from the torchdata repo. We’ll also be revisiting the DataPipes references in pytorch/pytorch. In release torchdata==0.8.0 (July 2024) they will be marked as deprecated, and in 0.9.0 (Oct 2024) they will be deleted. Existing users are advised to pin to torchdata==0.8.0 or an older version until they are able to migrate away. Subsequent releases will not include DataPipes or DataLoaderV2. Please reach out if you suggestions or comments (please use this issue for feedback)

ParquetDataFrameLoader

class torchdata.datapipes.iter.ParquetDataFrameLoader(source_dp: IterDataPipe[str], dtype=None, columns: Optional[List[str]] = None, device: str = '', use_threads: bool = False)

Takes in paths to Parquet files and return a TorchArrow DataFrame for each row group within a Parquet file (functional name: load_parquet_as_df).

Parameters:

• source_dp – source DataPipe containing paths to the Parquet files

• columns – List of str that specifies the column names of the DataFrame

• use_threads – if True, Parquet reader will perform multi-threaded column reads

• dtype – specify the TorchArrow dtype for the DataFrame, use torcharrow.dtypes.DType

• device – specify the device on which the DataFrame will be stored

Example

>>> from torchdata.datapipes.iter import FileLister
>>> import torcharrow.dtypes as dt
>>> DTYPE = dt.Struct([dt.Field("Values", dt.int32)])
>>> source_dp = FileLister(".", masks="df*.parquet")
>>> parquet_df_dp = source_dp.load_parquet_as_df(dtype=DTYPE)
>>> list(parquet_df_dp)[0]
  index    Values
-------  --------
      0         0
      1         1
      2         2
dtype: Struct([Field('Values', int32)]), count: 3, null_count: 0