org.dice_research.topicmodeling.algorithm.mallet

Class MalletLdaWrapper.MalletLDATopicModeler

java.lang.Object

cc.mallet.topics.ParallelTopicModel

org.dice_research.topicmodeling.algorithm.mallet.MalletLdaWrapper.MalletLDATopicModeler

All Implemented Interfaces:

Serializable, org.dice_research.topicmodeling.algorithms.ClassificationModel, org.dice_research.topicmodeling.algorithms.LDAModel, org.dice_research.topicmodeling.algorithms.Model, org.dice_research.topicmodeling.algorithms.ProbabilisticWordTopicModel, org.dice_research.topicmodeling.algorithms.VocabularyContaining, org.dice_research.topicmodeling.algorithms.VocabularyContainingClassificationModel, org.dice_research.topicmodeling.algorithms.VocabularyContainingModel

Enclosing class:

MalletLdaWrapper

protected static class MalletLdaWrapper.MalletLDATopicModeler
extends cc.mallet.topics.ParallelTopicModel
implements org.dice_research.topicmodeling.algorithms.LDAModel

See Also:

Serialized Form

Field Summary

Fields

Modifier and Type	Field and Description
protected MalletLdaInferenceWrapper	inferencer
protected int	inferencerVersion
protected int	iteration
protected static org.slf4j.Logger	logger
protected cc.mallet.topics.WorkerRunnable[]	runnables
private static long	serialVersionUID
protected double[]	topicWeights
protected org.dice_research.topicmodeling.utils.vocabulary.Vocabulary	vocabulary
protected double[][]	wordTopicWeights

Fields inherited from class cc.mallet.topics.ParallelTopicModel

alpha, alphabet, alphaSum, beta, betaSum, burninPeriod, data, DEFAULT_BETA, docLengthCounts, formatter, modelFilename, numIterations, numTopics, numTypes, optimizeInterval, printLogLikelihood, randomSeed, saveModelInterval, saveSampleInterval, saveStateInterval, showTopicsInterval, stateFilename, temperingInterval, tokensPerTopic, topicAlphabet, topicBits, topicDocCounts, topicMask, totalTokens, typeTopicCounts, UNASSIGNED_TOPIC, usingSymmetricAlpha, wordsPerTopic

Constructor Summary

Constructors

Constructor and Description
MalletLDATopicModeler(int numberOfTopics, double alphaSum, double beta, long seed)
MalletLDATopicModeler(int numberOfTopics, long seed)
MalletLDATopicModeler(cc.mallet.types.LabelAlphabet topicAlphabet, double alphaSum, double beta, long seed)

Method Summary

Modifier and Type	Method and Description
protected void	calculateSmoothedWeights()
void	estimate()
double[]	getAlphas()
double	getBeta()
org.dice_research.topicmodeling.utils.doc.DocumentClassificationResult	getClassificationForDocument(org.dice_research.topicmodeling.utils.doc.Document document)
cc.mallet.topics.TopicInferencer	getInferencer()
int	getNumberOfTopics()
double	getProbabilityOfWord(int wordId, int topicId)
double	getSmoothedProbabilityOfTopic(int topicId)
double	getSmoothedProbabilityOfWord(int wordId, int topicId)
double[]	getTopicProbabilitiesForDocument(org.dice_research.topicmodeling.utils.doc.DocumentWordCounts wordCounts)
int	getVersion()
org.dice_research.topicmodeling.utils.vocabulary.Vocabulary	getVocabulary()
org.dice_research.topicmodeling.utils.vocabulary.VocabularyMapping	getVocabularyMapping(org.dice_research.topicmodeling.utils.vocabulary.Vocabulary otherVocabulary)
int[]	inferTopicAssignmentsForDocument(org.dice_research.topicmodeling.utils.doc.Document document)
int[]	inferTopicAssignmentsForDocument(org.dice_research.topicmodeling.utils.doc.DocumentWordCounts wordCounts)
int[]	inferTopicAssignmentsForDocument(int[] tokens)
void	initialize(cc.mallet.types.InstanceList instances)
void	setVersion(int version)
void	setVocabularyDecorator(org.dice_research.topicmodeling.utils.vocabulary.VocabularyDecorator vocabulary)

Methods inherited from class cc.mallet.topics.ParallelTopicModel

addInstances, buildInitialTypeTopicCounts, displayTopWords, getAlphabet, getData, getNumTopics, getProbEstimator, getSortedWords, getTopicAlphabet, getTopicProbabilities, getTopicProbabilities, getTopWords, initializeFromState, main, modelLogLikelihood, optimizeAlpha, optimizeBeta, printDocumentTopics, printDocumentTopics, printDocumentTopics, printState, printState, printTopicWordWeights, printTopicWordWeights, printTopWords, printTopWords, printTypeTopicCounts, read, setBurninPeriod, setNumIterations, setNumThreads, setOptimizeInterval, setRandomSeed, setSaveSerializedModel, setSaveState, setSymmetricAlpha, setTemperingInterval, setTopicDisplay, sumTypeTopicCounts, temperAlpha, topicPhraseXMLReport, topicXMLReport, write

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

logger

protected static final org.slf4j.Logger logger

runnables

protected transient cc.mallet.topics.WorkerRunnable[] runnables

iteration

protected int iteration

inferencerVersion

protected transient int inferencerVersion

inferencer

protected transient MalletLdaInferenceWrapper inferencer

vocabulary

protected org.dice_research.topicmodeling.utils.vocabulary.Vocabulary vocabulary

wordTopicWeights

protected double[][] wordTopicWeights

topicWeights

protected double[] topicWeights

Constructor Detail

MalletLDATopicModeler

public MalletLDATopicModeler(int numberOfTopics,
                             long seed)

MalletLDATopicModeler

public MalletLDATopicModeler(int numberOfTopics,
                             double alphaSum,
                             double beta,
                             long seed)

MalletLDATopicModeler

public MalletLDATopicModeler(cc.mallet.types.LabelAlphabet topicAlphabet,
                             double alphaSum,
                             double beta,
                             long seed)

Method Detail

initialize

public void initialize(cc.mallet.types.InstanceList instances)

estimate

public void estimate()

Overrides:

estimate in class cc.mallet.topics.ParallelTopicModel

getSmoothedProbabilityOfWord

public double getSmoothedProbabilityOfWord(int wordId,
                                           int topicId)

Specified by:

getSmoothedProbabilityOfWord in interface org.dice_research.topicmodeling.algorithms.ProbabilisticWordTopicModel

getProbabilityOfWord

public double getProbabilityOfWord(int wordId,
                                   int topicId)

Specified by:

getProbabilityOfWord in interface org.dice_research.topicmodeling.algorithms.ProbabilisticWordTopicModel

getSmoothedProbabilityOfTopic

public double getSmoothedProbabilityOfTopic(int topicId)

Specified by:

getSmoothedProbabilityOfTopic in interface org.dice_research.topicmodeling.algorithms.ProbabilisticWordTopicModel

getNumberOfTopics

public int getNumberOfTopics()

Specified by:

getNumberOfTopics in interface org.dice_research.topicmodeling.algorithms.ProbabilisticWordTopicModel

calculateSmoothedWeights

protected void calculateSmoothedWeights()

getTopicProbabilitiesForDocument

public double[] getTopicProbabilitiesForDocument(org.dice_research.topicmodeling.utils.doc.DocumentWordCounts wordCounts)

Specified by:

getTopicProbabilitiesForDocument in interface org.dice_research.topicmodeling.algorithms.ProbabilisticWordTopicModel

getVocabulary

public org.dice_research.topicmodeling.utils.vocabulary.Vocabulary getVocabulary()

Specified by:

getVocabulary in interface org.dice_research.topicmodeling.algorithms.VocabularyContaining

getInferencer

public cc.mallet.topics.TopicInferencer getInferencer()

Overrides:

getInferencer in class cc.mallet.topics.ParallelTopicModel

setVocabularyDecorator

public void setVocabularyDecorator(org.dice_research.topicmodeling.utils.vocabulary.VocabularyDecorator vocabulary)

setVersion

public void setVersion(int version)

Specified by:

setVersion in interface org.dice_research.topicmodeling.algorithms.Model

getVersion

public int getVersion()

Specified by:

getVersion in interface org.dice_research.topicmodeling.algorithms.Model

getVocabularyMapping

public org.dice_research.topicmodeling.utils.vocabulary.VocabularyMapping getVocabularyMapping(org.dice_research.topicmodeling.utils.vocabulary.Vocabulary otherVocabulary)

Specified by:

getVocabularyMapping in interface org.dice_research.topicmodeling.algorithms.VocabularyContainingModel

getClassificationForDocument

public org.dice_research.topicmodeling.utils.doc.DocumentClassificationResult getClassificationForDocument(org.dice_research.topicmodeling.utils.doc.Document document)

Specified by:

getClassificationForDocument in interface org.dice_research.topicmodeling.algorithms.ClassificationModel

inferTopicAssignmentsForDocument

public int[] inferTopicAssignmentsForDocument(org.dice_research.topicmodeling.utils.doc.Document document)

Specified by:

inferTopicAssignmentsForDocument in interface org.dice_research.topicmodeling.algorithms.LDAModel

inferTopicAssignmentsForDocument

public int[] inferTopicAssignmentsForDocument(org.dice_research.topicmodeling.utils.doc.DocumentWordCounts wordCounts)

Specified by:

inferTopicAssignmentsForDocument in interface org.dice_research.topicmodeling.algorithms.LDAModel

inferTopicAssignmentsForDocument

public int[] inferTopicAssignmentsForDocument(int[] tokens)

Specified by:

inferTopicAssignmentsForDocument in interface org.dice_research.topicmodeling.algorithms.LDAModel

getBeta

public double getBeta()

Specified by:

getBeta in interface org.dice_research.topicmodeling.algorithms.LDAModel

getAlphas

public double[] getAlphas()

Specified by:

getAlphas in interface org.dice_research.topicmodeling.algorithms.LDAModel