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)

Saver

class torchdata.datapipes.iter.Saver(source_datapipe: IterDataPipe[Tuple[Any, Union[bytes, bytearray, str]]], mode: str = 'w', filepath_fn: Optional[Callable] = None)

Takes in a DataPipe of tuples of metadata and data, saves the data to the target path generated by the filepath_fn and metadata, and yields file path on local file system (functional name: save_to_disk).

Parameters:

• source_datapipe – Iterable DataPipe with tuples of metadata and data

• mode – Node in which the file will be opened for write the data ("w" by default)

• filepath_fn – Function that takes in metadata and returns the target path of the new file

Example

>>> from torchdata.datapipes.iter import IterableWrapper
>>> import os
>>> def filepath_fn(name: str) -> str:
>>>     return os.path.join(".", os.path.basename(name))
>>> name_to_data = {"1.txt": b"DATA1", "2.txt": b"DATA2", "3.txt": b"DATA3"}
>>> source_dp = IterableWrapper(sorted(name_to_data.items()))
>>> saver_dp = source_dp.save_to_disk(filepath_fn=filepath_fn, mode="wb")
>>> res_file_paths = list(saver_dp)
>>> res_file_paths
['./1.txt', './2.txt', './3.txt']