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)

Flattener

class torchdata.datapipes.iter.Flattener(datapipe: IterDataPipe, indices: Optional[Union[Hashable, List[Hashable]]] = None)

returns a flattened copy of the input DataPipe at the per sample/element level based on provided indices (functional name: flatten).

Note

no args will flatten each item in the datapipe 1 level

Parameters:

• datapipe – IterDataPipe with iterable elements

• indices –

  a single index/key for the item to flatten from an iterator item or a list of indices/keys to be flattened

  • Integer(s) is/are used for list/tuple.

  • Key(s) is/are used for dict.

Example

>>> from torchdata.datapipes.iter import IterableWrapper
>>> dp = IterableWrapper([(0, 10, (100, 1000)), (1, 11, (111, 1001)), (2, 12, (122, 1002)), (3, 13, (133, 1003)), (4, 14, (144, 1004))])
>>> flatten_dp = dp.flatten(2)
>>> list(flatten_dp)
[(0, 10, 100, 1000), (1, 11, 111, 1001), (2, 12, 122, 1002), (3, 13, 133, 1003), (4, 14, 144, 1004)]
>>>
>>> dp = IterableWrapper([(0, (1, 2)), (3, (4, 5)), (6, (7, 8))])
>>> flatten_dp = dp.flatten()
>>> list(flatten_dp)
[(0, 1, 2), (3, 4, 5), (6, 7, 8)]