reference/generated/tensordict.TensorDictBase.cummin

Run in Google Colab

Colab

Download Notebook

Notebook

View on GitHub

GitHub

TensorDictBase.cummin

TensorDictBase.cummin(dim: int, *, reduce: bool | None = None, return_indices: bool = True) → Self | Tensor

Returns the cumulative minimum of every leaf along a batch dimension.

As with torch.cummin(), the result is a named tuple of two tensordicts, values and indices, with the structure and the batch size of the tensordict.

Parameters:

dim (int) – the batch dimension along which to compute it. A negative dim counts from the end of the batch dimensions. A tuple uses only its first dimension; passing one is deprecated and will be removed in TensorDict 0.17.

Keyword Arguments:

• reduce (bool, optional) – if True, concatenates the leaves along dim, which requires them to have the same number of dimensions, and returns what torch.cummin() returns for the result. With feature dimensions, a negative dim counts from the end of each leaf. Starting with TensorDict 0.17, the default will change to counting it from the end of the batch dimensions, as with reduce=False. Defaults to False.

• return_indices (bool, optional) – if False, returns only the values tensordict. Defaults to True.

Examples

>>> import torch
>>> from tensordict import TensorDict
>>> td = TensorDict(a=torch.tensor([3.0, 1.0, 2.0]), batch_size=[3])
>>> td.cummin(dim=0).values["a"]
tensor([3., 1., 1.])
>>> td.cummin(dim=0).indices["a"]
tensor([0, 1, 1])

Every tensordict class and every tensorclass has this method. For the other methods, see Methods by task.