C# Class Accord.Statistics.Links.ProbitLinkFunction

Inheritance: ILinkFunction

Show file Open project: accord-net/framework Class Usage Examples

Public Methods

Method	Description
Clone ( ) : object	Creates a new object that is a copy of the current instance.
Derivative ( double x ) : double	First derivative of the Inverse function. The first derivative of the identity link function is given by `f'(x) = exp(c - (Phi^-1(x))² * 0.5)` in which `c = -log(sqrt(2*π)` and Phi^-1 is the inverse Normal (Gaussian) cumulative distribution function.
Derivative2 ( double y ) : double	First derivative of the Inverse function expressed in terms of it's output. The first derivative of the identity link function in terms of y = f(x) is given by `f'(y) = exp(c - x * x * 0.5)` in which `c = -log(sqrt(2*π)` and `x =`
Function ( double x ) : double	The Probit link function. The Probit link function is given by `f(x) = Phi^-1(x)`, in which Phi^-1 is the inverse Normal (Gaussian) cumulative distribution function.
Inverse ( double x ) : double	The Probit mean (activation) function. The Probit link function is given by `g(x) = Phi(x)`, in which Phi is the Normal (Gaussian) cumulative distribution function.
Log ( double x ) : double	The logarithm of the inverse of the link function.
ProbitLinkFunction ( ) : System	Creates a new Probit link function.

Method Details

Clone() public method

Creates a new object that is a copy of the current instance.

public Clone ( ) : object
return	object

Derivative() public method

First derivative of the Inverse function.

The first derivative of the identity link function is given by f'(x) = exp(c - (Phi^-1(x))² * 0.5) in which c = -log(sqrt(2*π) and Phi^-1 is the inverse Normal (Gaussian) cumulative distribution function.

public Derivative ( double x ) : double
x	double	The input value.
return	double

Derivative2() public method

First derivative of the Inverse function expressed in terms of it's output.

The first derivative of the identity link function in terms of y = f(x) is given by f'(y) = exp(c - x * x * 0.5) in which c = -log(sqrt(2*π) and x =

public Derivative2 ( double y ) : double
y	double	The reverse transformed value.
return	double

Function() public method

The Probit link function.

The Probit link function is given by f(x) = Phi^-1(x), in which Phi^-1 is the inverse Normal (Gaussian) cumulative distribution function.

public Function ( double x ) : double
x	double	An input value.
return	double

Inverse() public method

The Probit mean (activation) function.

The Probit link function is given by g(x) = Phi(x), in which Phi is the Normal (Gaussian) cumulative distribution function.

public Inverse ( double x ) : double
x	double	A transformed value.
return	double

Log() public method

The logarithm of the inverse of the link function.

public Log ( double x ) : double
x	double	A transformed value.
return	double

ProbitLinkFunction() public method

Creates a new Probit link function.

public ProbitLinkFunction ( ) : System
return	System