reference/generated/tensordict.TensorDictBase.cummax

Run in Google Colab

Colab

Download Notebook

Notebook

View on GitHub

GitHub

TensorDictBase.cummax

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

Returns the cumulative maximum values of all elements in the input tensordict.

Parameters:

dim (int) – integer representing the dimension along which to perform the cummax operation. A tuple uses only its first dim; passing one is deprecated and will be removed in TensorDict 0.17.

Keyword Arguments:

• reduce (bool, optional) – if True, the reduction will occur across all TensorDict values and a single reduced tensor will be returned. Defaults to False. With feature dims, 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 dims, as with reduce=False.

• return_indices (bool, optional) – cummax() returns a named tuple with values and indices when the dim argument is passed. The TensorDict equivalent of this is to return a named tuple with values and indices tensordicts of identical structure. If False, only the values are returned. Defaults to True.

Examples

>>> from tensordict import TensorDict
>>> import torch
>>> _ = torch.manual_seed(0)
>>> td = TensorDict(
...     a=torch.randn(3, 4, 5),
...     b=TensorDict(
...         c=torch.randn(3, 4, 5, 6),
...         d=torch.randn(3, 4, 5),
...         batch_size=(3, 4, 5),
...     ),
...     batch_size=(3, 4)
... )
>>> td.cummax(dim=0)
torch.return_types.cummax(
values=TensorDict(
    fields={
        a: Tensor(shape=torch.Size([3, 4, 5]), device=cpu, dtype=torch.float32, is_shared=False),
        b: TensorDict(
            fields={
                c: Tensor(shape=torch.Size([3, 4, 5, 6]), device=cpu, dtype=torch.float32, is_shared=False),
                d: Tensor(shape=torch.Size([3, 4, 5]), device=cpu, dtype=torch.float32, is_shared=False)},
            batch_size=torch.Size([3, 4]),
            device=None,
            is_shared=False)},
    batch_size=torch.Size([3, 4]),
    device=None,
    is_shared=False),
indices=TensorDict(
    fields={
        a: Tensor(shape=torch.Size([3, 4, 5]), device=cpu, dtype=torch.int64, is_shared=False),
        b: TensorDict(
            fields={
                c: Tensor(shape=torch.Size([3, 4, 5, 6]), device=cpu, dtype=torch.int64, is_shared=False),
                d: Tensor(shape=torch.Size([3, 4, 5]), device=cpu, dtype=torch.int64, is_shared=False)},
            batch_size=torch.Size([3, 4]),
            device=None,
            is_shared=False)},
    batch_size=torch.Size([3, 4]),
    device=None,
    is_shared=False))
>>> td = TensorDict(
...     a=torch.randn(3, 4, 5),
...     b=TensorDict(
...         c=torch.randn(3, 4, 5),
...         d=torch.randn(3, 4, 5),
...         batch_size=(3, 4, 5),
...     ),
...     batch_size=(3, 4)
... )
>>> td.cummax(reduce=True, dim=0)
torch.return_types.cummax(...)

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