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)

MapKeyZipper

class torchdata.datapipes.iter.MapKeyZipper(source_iterdatapipe: IterDataPipe, map_datapipe: MapDataPipe, key_fn: Callable, merge_fn: Optional[Callable] = None, keep_key: bool = False)

Joins the items from the source IterDataPipe with items from a MapDataPipe (functional name: zip_with_map). The matching is done by the provided key_fn, which maps an item from source_iterdatapipe to a key that should exist in the map_datapipe. The return value is created by the merge_fn, which returns a tuple of the two items by default.

Parameters:

• source_iterdatapipe – IterDataPipe from which items are yield and will be combined with an item from map_datapipe

• map_datapipe – MapDataPipe that takes a key from key_fn, and returns an item

• key_fn – Function that maps each item from source_iterdatapipe to a key that exists in map_datapipe

• keep_key – Option to yield the matching key along with the items in a tuple, resulting in (key, merge_fn(item1, item2)).

• merge_fn – Function that combines the item from source_iterdatapipe and the matching item from map_datapipe, by default a tuple is created

Example:

from torchdata.datapipes.iter import IterableWrapper
from torchdata.datapipes.map import SequenceWrapper


def merge_fn(tuple_from_iter, value_from_map):
    return tuple_from_iter[0], tuple_from_iter[1] + value_from_map


dp1 = IterableWrapper([('a', 1), ('b', 2), ('c', 3)])
mapdp = SequenceWrapper({'a': 100, 'b': 200, 'c': 300, 'd': 400})
res_dp = dp1.zip_with_map(map_datapipe=mapdp, key_fn=itemgetter(0), merge_fn=merge_fn)

print(list(res_dp))

Output:

[('a', 101), ('b', 202), ('c', 303)]