log_softmax(input, dim=None, _stacklevel=3, dtype=None)¶
Applies a softmax followed by a logarithm.
While mathematically equivalent to log(softmax(x)), doing these two operations separately is slower, and numerically unstable. This function uses an alternative formulation to compute the output and gradient correctly.
LogSoftmaxfor more details.
input (Tensor) – input
dim (int) – A dimension along which log_softmax will be computed.
torch.dtype, optional) – the desired data type of returned tensor. If specified, the input tensor is casted to
dtypebefore the operation is performed. This is useful for preventing data type overflows. Default: None.