org.openimaj.pgm.vb.lda.mle

Class LDALearner

java.lang.Object

org.openimaj.pgm.vb.lda.mle.LDALearner

public class LDALearner
extends Object

An implementation of variational inference LDA which can be saved and loaded

Author:

Sina Samangooei (ss@ecs.soton.ac.uk)

Constructor Summary

Constructors

Constructor and Description
LDALearner(int ntopics)

Method Summary

Modifier and Type	Method and Description
double	computeLikelihood(Document doc, LDAVariationlState vstate) Calculates a lower bound for the log liklihood of a document given current parameters.
void	estimate(Corpus corpus) initiates the EM algorithm on documents in the corpus
<T> T	getConfig(org.openimaj.pgm.vb.lda.mle.LDALearner.LDAConfig key)

Methods inherited from class java.lang.Object

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

Constructor Detail

LDALearner

public LDALearner(int ntopics)

Parameters:

ntopics -

Method Detail

getConfig

public <T> T getConfig(org.openimaj.pgm.vb.lda.mle.LDALearner.LDAConfig key)

Parameters:

key -

Returns:

the configuration parameter value

estimate

public void estimate(Corpus corpus)

initiates the EM algorithm on documents in the corpus

Parameters:

corpus -

computeLikelihood

public double computeLikelihood(Document doc,
                                LDAVariationlState vstate)

Calculates a lower bound for the log liklihood of a document given current parameters. When this is maximised it minimises the KL divergence between the the variation posterior and the true posterior. The derivation can be seen in the appendix of Blei's LDA paper 2003

Parameters:

doc -

vstate -

Returns:

the likelihood