softplus — SciPy v1.15.2 Manual (original) (raw)

scipy.special.

scipy.special.softplus(x, **kwargs)[source]#

Compute the softplus function element-wise.

The softplus function is defined as: softplus(x) = log(1 + exp(x)). It is a smooth approximation of the rectifier function (ReLU).

Parameters:

xarray_like

Input value.

**kwargs

For other keyword-only arguments, see theufunc docs.

Returns:

softplusndarray

Logarithm of exp(0) + exp(x).

Examples

from scipy import special

special.softplus(0) 0.6931471805599453

special.softplus([-1, 0, 1]) array([0.31326169, 0.69314718, 1.31326169])