name: Sigmoid


g++ [flags ...] file ... -l /isip/tools/lib/$ISIP_BINARY/lib_numeric.a

#include <Sigmoid.h>

boolean set(float gain, float slope, float xoffset, float yoffset);
boolean compute(float& y, float x);
quick start:

Sigmoid sigm(1, 0.5);
float y;
float x = 0.5;

// evaluate y = sigmoid(x)
sigm.compute(y, x);

// evaluate y = derivative with respect to 'x' of the sigmoid function
// evaluated at x
sigm.derivative(y, x);

The Sigmoid class is used to analyze and evaluate sigmoid functions of the form: Nominally, the gain is 1, the offset is 0 and the slope is 1, which produces a function that varies smoothly between 0 and 1. The gain, x_offset and y_offset, can be adjusted to control the effective domain and range of the output. The variable named slope is not truly the slope of the transition region but can be varied to control that slope.


public constants:

error codes:

protected data:

required public methods:

class-specific public methods:

private methods: