jax.nn.log_softmax(x, axis=-1, where=None, initial=None)[source]#

Log-Softmax function.

Computes the logarithm of the softmax function, which rescales elements to the range \([-\infty, 0)\).

\[\mathrm{log\_softmax}(x) = \log \left( \frac{\exp(x_i)}{\sum_j \exp(x_j)} \right)\]
  • x (Any) – input array

  • axis (Union[int, Tuple[int, ...], None]) – the axis or axes along which the log_softmax should be computed. Either an integer or a tuple of integers.

  • where (Optional[Any]) – Elements to include in the log_softmax.

  • initial (Optional[Any]) – The minimum value used to shift the input array. Must be present when where is not None.

Return type: