org.knowceans.dirichlet.lda
Class LdaConfiguration

java.lang.Object
  org.knowceans.dirichlet.lda.LdaConfiguration

All Implemented Interfaces:

java.io.Serializable

Direct Known Subclasses:

ExtLdaConfiguration

public class LdaConfiguration
extends java.lang.Object
implements java.io.Serializable

LdaConfiguration handles the configuration of an lda gibbs sampler. The fields are public to save the time for getter method calls in inner loops.

Author:

gregor

See Also:

Serialized Form

Field Summary

double	alpha Dirichlet parameter (document--topic associations)
double	beta Dirichlet parameter (topic--term associations)
int	burnIn burn-in period
int	iterations max iterations
int	K number of topics
int	sampleLag sample lag (if -1 only one sample taken)
private static long	serialVersionUID
int	thinInterval sampling lag (?)

Constructor Summary

LdaConfiguration()
Read lda configuration from conf file.

LdaConfiguration(int iterations, int burnIn, int thinInterval, int sampleLag, int K, double alpha, double beta)
Configure the gibbs sampler

Method Summary

Methods inherited from class java.lang.Object

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

Field Detail

serialVersionUID

private static final long serialVersionUID

See Also:

Constant Field Values

K

public int K

number of topics

alpha

public double alpha

Dirichlet parameter (document--topic associations)

beta

public double beta

Dirichlet parameter (topic--term associations)

burnIn

public int burnIn

burn-in period

iterations

public int iterations

max iterations

sampleLag

public int sampleLag

sample lag (if -1 only one sample taken)

thinInterval

public int thinInterval

sampling lag (?)

Constructor Detail

LdaConfiguration

public LdaConfiguration(int iterations,
                        int burnIn,
                        int thinInterval,
                        int sampleLag,
                        int K,
                        double alpha,
                        double beta)

Configure the gibbs sampler

Parameters:

iterations - number of total iterations

burnIn - number of burn-in iterations

thinInterval - update statistics interval

sampleLag - sample interval (-1 for just one sample at the end)

K - number of topics

alpha - symmetric prior parameter on document--topic associations

beta - symmetric prior parameter on topic--term associations

LdaConfiguration

public LdaConfiguration()

Read lda configuration from conf file.