log_softmax¶

dragon.vm.torch.nn.functional.log_softmax(
  input,
  dim,
  inplace=False
)[source]¶

Apply the composite of logarithm and softmax to input.

The LogSoftmax function is defined as:

\[\text{LogSoftmax}(x_{i}) = \log(\frac{\exp(x_{i})}{\sum \exp(x_{j})}) \]

Parameters:

• input (dragon.vm.torch.Tensor) – The input.
• dim (int) – The dimension to reduce.
• inplace (bool, optional, default=False) – Whether to do the operation in-place.

Returns:

dragon.vm.torch.Tensor – The output tensor.

See also

torch.nn.LogSoftmax(…)