Slicer

class torchdata.datapipes.iter.Slicer(datapipe: IterDataPipe, index: Union[int, List[Hashable]], stop: Optional[int] = None, step: Optional[int] = None)

returns a slice of elements in input DataPipe via start/stop/step or indices (functional name: slice).

Parameters:

• datapipe – IterDataPipe with iterable elements

• index –

  a single start index for the slice or a list of indices to be returned instead of a start/stop slice

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

  • Key(s) is/are used for dict.

• stop – the slice stop. ignored if index is a list or if element is a dict

• step – step to be taken from start to stop. ignored if index is a list or if element is a dict

Example

>>> from torchdata.datapipes.iter import IterableWrapper
>>> dp = IterableWrapper([(0, 10, 100), (1, 11, 111), (2, 12, 122), (3, 13, 133), (4, 14, 144)])
>>> slice_dp = dp.slice(0, 2)
>>> list(slice_dp)
[(0, 10), (1, 11), (2, 12), (3, 13), (4, 14)]