Softmin — nnf_softmin • torch

Applies a softmin function.

Usage

nnf_softmin(input, dim, dtype = NULL)

input: (Tensor) input
dim: (int) A dimension along which softmin will be computed (so every slice along dim will sum to 1).
dtype: (torch.dtype, optional) the desired data type of returned tensor. If specified, the input tensor is casted to dtype before the operation is performed. This is useful for preventing data type overflows. Default: NULL.

Note that

$$Softmin(x) = Softmax(-x)$$.

See nnf_softmax definition for mathematical formula.