Package cc.mallet.topics

Class LDAHyper

  • All Implemented Interfaces:
    java.io.Serializable
    Direct Known Subclasses:
    LDAStream
    @Deprecated
public class LDAHyper
extends java.lang.Object
implements java.io.Serializable
    Deprecated.
    Use ParallelTopicModel instead, which uses substantially faster data structures even for non-parallel operation.
    Latent Dirichlet Allocation with optimized hyperparameters
    Author:
    David Mimno, Andrew McCallum
    See Also:
    Serialized Form

    • Field Detail

      • alphabet

        protected Alphabet alphabet
        Deprecated.

      • topicAlphabet

        protected LabelAlphabet topicAlphabet
        Deprecated.

      • numTopics

        protected int numTopics
        Deprecated.

      • numTypes

        protected int numTypes
        Deprecated.

      • alpha

        protected double[] alpha
        Deprecated.

      • alphaSum

        protected double alphaSum
        Deprecated.

      • beta

        protected double beta
        Deprecated.

      • betaSum

        protected double betaSum
        Deprecated.

      • smoothingOnlyMass

        protected double smoothingOnlyMass
        Deprecated.

      • cachedCoefficients

        protected double[] cachedCoefficients
        Deprecated.

      • oneDocTopicCounts

        protected int[] oneDocTopicCounts
        Deprecated.

      • typeTopicCounts

        protected com.carrotsearch.hppc.IntIntHashMap[] typeTopicCounts
        Deprecated.

      • tokensPerTopic

        protected int[] tokensPerTopic
        Deprecated.

      • docLengthCounts

        protected int[] docLengthCounts
        Deprecated.

      • topicDocCounts

        protected int[][] topicDocCounts
        Deprecated.

      • iterationsSoFar

        public int iterationsSoFar
        Deprecated.

      • numIterations

        public int numIterations
        Deprecated.

      • burninPeriod

        public int burninPeriod
        Deprecated.

      • saveSampleInterval

        public int saveSampleInterval
        Deprecated.

      • optimizeInterval

        public int optimizeInterval
        Deprecated.

      • showTopicsInterval

        public int showTopicsInterval
        Deprecated.

      • wordsPerTopic

        public int wordsPerTopic
        Deprecated.

      • outputModelInterval

        protected int outputModelInterval
        Deprecated.

      • outputModelFilename

        protected java.lang.String outputModelFilename
        Deprecated.

      • saveStateInterval

        protected int saveStateInterval
        Deprecated.

      • stateFilename

        protected java.lang.String stateFilename
        Deprecated.

      • random

        protected Randoms random
        Deprecated.

      • formatter

        protected java.text.NumberFormat formatter
        Deprecated.

      • printLogLikelihood

        protected boolean printLogLikelihood
        Deprecated.

    • Constructor Detail

      • LDAHyper

        public LDAHyper​(int numberOfTopics)
        Deprecated.

      • LDAHyper

        public LDAHyper​(int numberOfTopics,
                double alphaSum,
                double beta)
        Deprecated.

      • LDAHyper

        public LDAHyper​(int numberOfTopics,
                double alphaSum,
                double beta,
                Randoms random)
        Deprecated.

      • LDAHyper

        public LDAHyper​(LabelAlphabet topicAlphabet,
                double alphaSum,
                double beta,
                Randoms random)
        Deprecated.

    • Method Detail

      • getAlphabet

        public Alphabet getAlphabet()
        Deprecated.

      • getTopicAlphabet

        public LabelAlphabet getTopicAlphabet()
        Deprecated.

      • getNumTopics

        public int getNumTopics()
        Deprecated.

      • getCountFeatureTopic

        public int getCountFeatureTopic​(int featureIndex,
                                int topicIndex)
        Deprecated.

      • getCountTokensPerTopic

        public int getCountTokensPerTopic​(int topicIndex)
        Deprecated.

      • setTestingInstances

        public void setTestingInstances​(InstanceList testing)
        Deprecated.
        Held-out instances for empirical likelihood calculation

      • setNumIterations

        public void setNumIterations​(int numIterations)
        Deprecated.

      • setBurninPeriod

        public void setBurninPeriod​(int burninPeriod)
        Deprecated.

      • setTopicDisplay

        public void setTopicDisplay​(int interval,
                            int n)
        Deprecated.

      • setRandomSeed

        public void setRandomSeed​(int seed)
        Deprecated.

      • setOptimizeInterval

        public void setOptimizeInterval​(int interval)
        Deprecated.

      • setModelOutput

        public void setModelOutput​(int interval,
                           java.lang.String filename)
        Deprecated.

      • setSaveState

        public void setSaveState​(int interval,
                         java.lang.String filename)
        Deprecated.
        Define how often and where to save the state
        Parameters:
        interval - Save a copy of the state every interval iterations.
        filename - Save the state to this file, with the iteration number as a suffix

      • instanceLength

        protected int instanceLength​(Instance instance)
        Deprecated.

      • addInstances

        public void addInstances​(InstanceList training)
        Deprecated.

      • initializeHistogramsAndCachedValues

        protected void initializeHistogramsAndCachedValues()
        Deprecated.
        Gather statistics on the size of documents and create histograms for use in Dirichlet hyperparameter optimization.

      • estimate

        public void estimate()
              throws java.io.IOException
        Deprecated.
        Throws:
        java.io.IOException

      • estimate

        public void estimate​(int iterationsThisRound)
              throws java.io.IOException
        Deprecated.
        Throws:
        java.io.IOException

      • sampleTopicsForOneDoc

        protected void sampleTopicsForOneDoc​(FeatureSequence tokenSequence,
                                     FeatureSequence topicSequence,
                                     boolean shouldSaveState,
                                     boolean readjustTopicsAndStats)
        Deprecated.

      • getSortedTopicWords

        public IDSorter[] getSortedTopicWords​(int topic)
        Deprecated.

      • printTopWords

        public void printTopWords​(java.io.File file,
                          int numWords,
                          boolean useNewLines)
                   throws java.io.IOException
        Deprecated.
        Throws:
        java.io.IOException

      • printTopWords

        public void printTopWords​(java.io.PrintStream out,
                          int numWords,
                          boolean usingNewLines)
        Deprecated.

      • topicXMLReport

        public void topicXMLReport​(java.io.PrintWriter out,
                           int numWords)
        Deprecated.

      • topicXMLReportPhrases

        public void topicXMLReportPhrases​(java.io.PrintStream out,
                                  int numWords)
        Deprecated.

      • printDocumentTopics

        public void printDocumentTopics​(java.io.File f)
                         throws java.io.IOException
        Deprecated.
        Throws:
        java.io.IOException

      • printDocumentTopics

        public void printDocumentTopics​(java.io.PrintWriter pw)
        Deprecated.

      • printDocumentTopics

        public void printDocumentTopics​(java.io.PrintWriter pw,
                                double threshold,
                                int max)
        Deprecated.
        Parameters:
        pw - A print writer
        threshold - Only print topics with proportion greater than this number
        max - Print no more than this many topics

      • printState

        public void printState​(java.io.File f)
                throws java.io.IOException
        Deprecated.
        Throws:
        java.io.IOException

      • printState

        public void printState​(java.io.PrintStream out)
        Deprecated.

      • write

        public void write​(java.io.File f)
        Deprecated.

      • read

        public static LDAHyper read​(java.io.File f)
        Deprecated.

      • topicLabelMutualInformation

        public double topicLabelMutualInformation()
        Deprecated.

      • empiricalLikelihood

        public double empiricalLikelihood​(int numSamples,
                                  InstanceList testing)
        Deprecated.

      • modelLogLikelihood

        public double modelLogLikelihood()
        Deprecated.

      • main

        public static void main​(java.lang.String[] args)
                 throws java.io.IOException
        Deprecated.
        Throws:
        java.io.IOException