reference/generated/torchrl.modules.functional.signed_parabolic

Run in Google Colab

Colab

Download Notebook

Notebook

View on GitHub

GitHub

signed_parabolic

class torchrl.modules.functional.signed_parabolic(value: Tensor, epsilon: float = 0.001)

Apply the inverse of signed_hyperbolic() element-wise.

Parameters:

• value (torch.Tensor) – Input tensor in signed-hyperbolic space.

• epsilon (float, optional) – Positive linear correction used by the corresponding signed_hyperbolic() call. Defaults to 1e-3.

Returns:

A tensor with the same shape, dtype, and device as value.

Examples

>>> import torch
>>> from torchrl.modules import functional as F
>>> value = torch.tensor([-100.0, 0.0, 100.0])
>>> transformed = F.signed_hyperbolic(value)
>>> transformed
tensor([-9.1499,  0.0000,  9.1499])
>>> F.signed_parabolic(transformed)
tensor([-100.0000,    0.0000,  100.0000])

See also

signed_hyperbolic() for the forward operation and SignedHyperbolicValueTransform for the module form.