softmax: Softmax Function with Special Handling for Infinite Values

In covalchemy: Constructing Joint Distributions with Control Over Statistical Properties

Softmax Function with Special Handling for Infinite Values

Description

This function computes the softmax of a vector x, with special handling for infinite values. The softmax function transforms input values into a probability distribution by exponentiating each value, then normalizing by the sum of all exponentiated values. The function ensures numerical stability, particularly when dealing with very large or very small values, and handles cases where the values are infinite (Inf).

Usage

softmax(x)

Arguments

x	A numeric vector for which the softmax function will be calculated.

Value

A numeric vector of the same length as x, where the values represent probabilities summing to 1.

Examples

softmax(c(10, 5, 2))
softmax(c(Inf, -Inf, 0))
softmax(c(-Inf, -Inf, -Inf))

covalchemy documentation built on April 12, 2025, 2:15 a.m.