torchact.nn.LogSoftmax

class torchact.nn.LogSoftmax(dim: Optional[int] = None)[source]

Implementation of Log(Softmax(x)).

Parameters: dim (int) – LogSoftmax dimension. Default: None

Examples::

>>> import torch, torchact
>>> m = torchact.nn.LogSoftmax()
>>> input = torch.tensor([1.0, -2.0, 0.0, 3.0])
>>> output = m(input)
>>> print(output)
tensor([-2.1755, -5.1755, -3.1755, -0.1755])