reference/generated/torchrl.modules.IdentityValueTransform

Run in Google Colab

Colab

Download Notebook

Notebook

View on GitHub

GitHub

IdentityValueTransform

class torchrl.modules.IdentityValueTransform(*args: Any, **kwargs: Any)

Leave scalar values unchanged.

Examples

>>> import torch
>>> from torchrl.modules import IdentityValueTransform
>>> transform = IdentityValueTransform()
>>> value = torch.tensor([-1.0, 0.0, 1.0])
>>> transformed = transform(value)
>>> transformed
tensor([-1.,  0.,  1.])
>>> transform.inverse(transformed)
tensor([-1.,  0.,  1.])

See also

ValueTransform for the interface and ComposeValueTransform for composing transforms.

forward(value: Tensor) → Tensor

Return value unchanged.

inverse(value: Tensor) → Tensor

Return value unchanged.