reference/generated/tensordict.nn.utils.expln

Run in Google Colab

Colab

Download Notebook

Notebook

View on GitHub

GitHub

expln

tensordict.nn.utils.expln(x)

Maps real numbers to positive numbers: exp(x) for x <= 0, log1p(x) + 1 for x > 0.

The function is smooth, and grows like a logarithm for large inputs. It comes from Rückstieß et al., “State-Dependent Exploration for Policy Gradient Methods” (ECML 2008, https://people.idsia.ch/~juergen/ecml2008rueckstiess.pdf). mappings() returns it for "expln".

Parameters:

x (Tensor) – the input.

Returns:

A new tensor with the shape of x.

Examples

>>> import torch
>>> from tensordict.nn.utils import expln
>>> expln(torch.tensor([-1.0, 0.0, 1.0]))
tensor([0.3679, 1.0000, 1.6931])