org.terrier.statistics
Class GammaFunction

java.lang.Object
  org.terrier.statistics.GammaFunction

All Implemented Interfaces:

java.io.Serializable

public abstract class GammaFunction
extends java.lang.Object
implements java.io.Serializable

Base class for implementations of the Gamma function. Use getGammaFunction() to obtain an instance. The exact instance can be controlled by property gamma.function For consistency when computing logarithms of the gamma function, all implementations assume positive input values. In practice, as the gamma function can generate large values, normal usage should use compute_log() anyway.

Properties:

<ul>

gamma.function - class name of the Gamma function implementation. Defaults to use WikipediaLanczosGammaFunction

Since:

3.0

Author:

Craig Macdonald

See Also:

Serialized Form

Constructor Summary

GammaFunction()

Method Summary

abstract double	compute_log(double number) Get the value of the log of gamma function for the specified number.
abstract double	compute(double number) Get the value of the gamma function for the specified number.
static long	factorial(long n) Compute factorial of n, for 0 < n < 21.
static GammaFunction	getGammaFunction() Obtain an instance of GammaFunction

Methods inherited from class java.lang.Object

clone, equals, finalize, getClass, hashCode, notify, notifyAll, toString, wait, wait, wait

Constructor Detail

GammaFunction

public GammaFunction()

Method Detail

compute

public abstract double compute(double number)

Get the value of the gamma function for the specified number.

Parameters:

number - for which is required

Returns:

(n-1)!

compute_log

public abstract double compute_log(double number)

Get the value of the log of gamma function for the specified number.

Parameters:

number - for which is required

Returns:

log(n-1)!

getGammaFunction

public static final GammaFunction getGammaFunction()

Obtain an instance of GammaFunction

factorial

public static final long factorial(long n)

Compute factorial of n, for 0 < n < 21.

Parameters:

n - number to compute for

Returns:

factorial of n