Dropper

class torchdata.datapipes.iter.Dropper(datapipe: IterDataPipe, indices: Union[Hashable, List[Hashable]])

Drop columns/elements in input DataPipe via its indices (functional name: drop).

Parameters:

• datapipe – IterDataPipe with columns to be dropped

• indices –

  a single column index to be dropped or a list of indices

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

  • Key(s) is/are used for dict.

Example

>>> from torchdata.datapipes.iter import IterableWrapper, ZipperMapDataPipe
>>> dp1 = IterableWrapper(range(5))
>>> dp2 = IterableWrapper(range(10, 15))
>>> dp = dp1.zip(dp2)
>>> list(dp)
[(0, 10), (1, 11), (2, 12), (3, 13), (4, 14)]
>>> drop_dp = dp.drop(1)
>>> list(drop_dp)
[(0), (1), (2), (3), (4)]