cc.mallet.topics
Class LDAStream

java.lang.Object
  extended by cc.mallet.topics.LDAHyper
      extended by cc.mallet.topics.LDAStream
All Implemented Interfaces:
java.io.Serializable

public class LDAStream
extends LDAHyper

Author:
Limin Yao, David Mimno
See Also:
Serialized Form

Nested Class Summary
 
Nested classes/interfaces inherited from class cc.mallet.topics.LDAHyper
LDAHyper.Topication
 
Field Summary
protected  java.util.ArrayList<LDAHyper.Topication> test
           
 
Fields inherited from class cc.mallet.topics.LDAHyper
alpha, alphabet, alphaSum, beta, betaSum, burninPeriod, cachedCoefficients, data, DEFAULT_BETA, docLengthCounts, formatter, iterationsSoFar, numIterations, numTopics, numTypes, oneDocTopicCounts, optimizeInterval, outputModelFilename, outputModelInterval, printLogLikelihood, random, saveSampleInterval, saveStateInterval, showTopicsInterval, smoothingOnlyMass, stateFilename, testing, tokensPerTopic, topicAlphabet, topicDocCounts, typeTopicCounts, wordsPerTopic
 
Constructor Summary
LDAStream(int numberOfTopics)
           
LDAStream(int numberOfTopics, double alphaSum, double beta)
           
LDAStream(int numberOfTopics, double alphaSum, double beta, Randoms random)
           
LDAStream(LabelAlphabet topicAlphabet, double alphaSum, double beta, Randoms random)
           
 
Method Summary
 void estimateAll(int iteration)
           
 java.util.ArrayList<LDAHyper.Topication> getTest()
           
 void inference(int maxIteration)
           
 void inferenceAll(int maxIteration)
           
 void inferenceOneByOne(int maxIteration)
           
 void inferenceWithTheta(int maxIteration, InstanceList theta)
           
 void printDocumentTopics(java.util.ArrayList<LDAHyper.Topication> dataset, java.io.File f)
           
 void printDocumentTopics(java.util.ArrayList<LDAHyper.Topication> dataset, java.io.PrintWriter pw)
           
 void printDocumentTopics(java.util.ArrayList<LDAHyper.Topication> dataset, java.io.PrintWriter pw, double threshold, int max)
           
 void printPhi(java.io.File f, double threshold)
           
 void printState(java.util.ArrayList<LDAHyper.Topication> dataset, java.io.File f)
           
 void printState(java.util.ArrayList<LDAHyper.Topication> dataset, java.io.PrintStream out)
           
 void printTheta(java.util.ArrayList<LDAHyper.Topication> dataset, java.io.File f, double threshold, int max)
           
 
Methods inherited from class cc.mallet.topics.LDAHyper
addInstances, addInstances, empiricalLikelihood, estimate, estimate, getAlphabet, getCountFeatureTopic, getCountTokensPerTopic, getData, getNumTopics, getSortedTopicWords, getTopicAlphabet, initializeHistogramsAndCachedValues, instanceLength, main, modelLogLikelihood, printDocumentTopics, printDocumentTopics, printDocumentTopics, printState, printState, printTopWords, printTopWords, read, sampleTopicsForOneDoc, setBurninPeriod, setModelOutput, setNumIterations, setOptimizeInterval, setRandomSeed, setSaveState, setTestingInstances, setTopicDisplay, topicLabelMutualInformation, topicXMLReport, topicXMLReportPhrases, write
 
Methods inherited from class java.lang.Object
clone, equals, finalize, getClass, hashCode, notify, notifyAll, toString, wait, wait, wait
 

Field Detail

test

protected java.util.ArrayList<LDAHyper.Topication> test
Constructor Detail

LDAStream

public LDAStream(int numberOfTopics)
Parameters:
numberOfTopics -

LDAStream

public LDAStream(int numberOfTopics,
                 double alphaSum,
                 double beta)
Parameters:
numberOfTopics -
alphaSum -
beta -

LDAStream

public LDAStream(int numberOfTopics,
                 double alphaSum,
                 double beta,
                 Randoms random)
Parameters:
numberOfTopics -
alphaSum -
beta -
random -

LDAStream

public LDAStream(LabelAlphabet topicAlphabet,
                 double alphaSum,
                 double beta,
                 Randoms random)
Parameters:
topicAlphabet -
alphaSum -
beta -
random -
Method Detail

getTest

public java.util.ArrayList<LDAHyper.Topication> getTest()

inferenceAll

public void inferenceAll(int maxIteration)

estimateAll

public void estimateAll(int iteration)
                 throws java.io.IOException
Throws:
java.io.IOException

inference

public void inference(int maxIteration)
Parameters:
maxIteration -
-

inferenceOneByOne

public void inferenceOneByOne(int maxIteration)

inferenceWithTheta

public void inferenceWithTheta(int maxIteration,
                               InstanceList theta)

printTheta

public void printTheta(java.util.ArrayList<LDAHyper.Topication> dataset,
                       java.io.File f,
                       double threshold,
                       int max)
                throws java.io.IOException
Throws:
java.io.IOException

printPhi

public void printPhi(java.io.File f,
                     double threshold)
              throws java.io.IOException
Throws:
java.io.IOException

printDocumentTopics

public void printDocumentTopics(java.util.ArrayList<LDAHyper.Topication> dataset,
                                java.io.File f)
                         throws java.io.IOException
Throws:
java.io.IOException

printDocumentTopics

public void printDocumentTopics(java.util.ArrayList<LDAHyper.Topication> dataset,
                                java.io.PrintWriter pw)

printDocumentTopics

public void printDocumentTopics(java.util.ArrayList<LDAHyper.Topication> dataset,
                                java.io.PrintWriter pw,
                                double threshold,
                                int max)
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.util.ArrayList<LDAHyper.Topication> dataset,
                       java.io.File f)
                throws java.io.IOException
Throws:
java.io.IOException

printState

public void printState(java.util.ArrayList<LDAHyper.Topication> dataset,
                       java.io.PrintStream out)