Package opennlp.tools.ml.model

Examples of opennlp.tools.ml.model.DataIndexer

64
65
66
67
68
69
70
71
72
73
74
    String dataIndexerName = getStringParam(DATA_INDEXER_PARAM,
        DATA_INDEXER_TWO_PASS_VALUE);

    int cutoff = getCutoff();
    boolean sortAndMerge = isSortAndMerge();
    DataIndexer indexer = null;

    if (DATA_INDEXER_ONE_PASS_VALUE.equals(dataIndexerName)) {
      indexer = new OnePassDataIndexer(events, cutoff, sortAndMerge);
    } else if (DATA_INDEXER_TWO_PASS_VALUE.equals(dataIndexerName)) {
      indexer = new TwoPassDataIndexer(events, cutoff, sortAndMerge);
View Full Code Here
86
87
88
89
90
91
92
93
94
95
96
    if (!isValid()) {
      throw new IllegalArgumentException("trainParams are not valid!");
    }

    HashSumEventStream hses = new HashSumEventStream(events);
    DataIndexer indexer = getDataIndexer(events);

    MaxentModel model = doTrain(indexer);

    addToReport("Training-Eventhash", hses.calculateHashSum().toString(16));
    addToReport(AbstractTrainer.TRAINER_TYPE_PARAM, EventTrainer.EVENT_VALUE);
View Full Code Here
116
117
118
119
120
121
122
123
124
125
126
127
128
129
130
131
132
133
134
135
136
137
138
139
140
141
142
143
144
145
146
147
148
149
  // << members related to AbstractSequenceTrainer

  public AbstractModel trainModel(int iterations, SequenceStream sequenceStream, int cutoff, boolean useAverage) throws IOException {
    this.iterations = iterations;
    this.sequenceStream = sequenceStream;
    DataIndexer di = new OnePassDataIndexer(new SequenceStreamEventStream(sequenceStream),cutoff,false);
    numSequences = 0;

    sequenceStream.reset();

    while (sequenceStream.read() != null) {
      numSequences++;
    }

    outcomeList  = di.getOutcomeList();
    predLabels = di.getPredLabels();
    pmap = new IndexHashTable<String>(predLabels, 0.7d);

    display("Incorporating indexed data for training...  \n");
    this.useAverage = useAverage;
    numEvents = di.getNumEvents();

    this.iterations = iterations;
    outcomeLabels = di.getOutcomeLabels();
    omap = new HashMap<String,Integer>();
    for (int oli=0;oli<outcomeLabels.length;oli++) {
      omap.put(outcomeLabels[oli], oli);
    }
    outcomeList = di.getOutcomeList();

    numPreds = predLabels.length;
    numOutcomes = outcomeLabels.length;
    if (useAverage) {
      updates = new int[numPreds][numOutcomes][3];
View Full Code Here
43
44
45
46
47
48
49
50
51
52
53
  @Test
  public void testTrainModelReturnsAQNModel() throws Exception {
    // given
    RealValueFileEventStream rvfes1 = new RealValueFileEventStream(
        "src/test/resources/data/opennlp/maxent/real-valued-weights-training-data.txt"); 
    DataIndexer testDataIndexer = new OnePassRealValueDataIndexer(rvfes1,1);
    // when
    QNModel trainedModel = new QNTrainer(false).trainModel(ITERATIONS, testDataIndexer);
    // then
    assertNotNull(trainedModel);
  }
View Full Code Here
55
56
57
58
59
60
61
62
63
64
65
  @Test
  public void testInTinyDevSet() throws Exception {
    // given
    RealValueFileEventStream rvfes1 = new RealValueFileEventStream(
        "src/test/resources/data/opennlp/maxent/real-valued-weights-training-data.txt"); 
    DataIndexer testDataIndexer = new OnePassRealValueDataIndexer(rvfes1,1);
    // when
    QNModel trainedModel = new QNTrainer(15, true).trainModel(ITERATIONS, testDataIndexer);
    String[] features2Classify = new String[] {
        "feature2","feature3", "feature3",
        "feature3","feature3", "feature3",
View Full Code Here
73
74
75
76
77
78
79
80
81
82
83
  @Test
  public void testModel() throws IOException {
      // given
      RealValueFileEventStream rvfes1 = new RealValueFileEventStream(
          "src/test/resources/data/opennlp/maxent/real-valued-weights-training-data.txt"); 
      DataIndexer testDataIndexer = new OnePassRealValueDataIndexer(rvfes1,1);
      // when
      QNModel trainedModel = new QNTrainer(15, true).trainModel(
          ITERATIONS, testDataIndexer);
     
      assertTrue(trainedModel.equals(trainedModel)); 
View Full Code Here
87
88
89
90
91
92
93
94
95
96
97
  @Test
  public void testSerdeModel() throws IOException {
      // given
      RealValueFileEventStream rvfes1 = new RealValueFileEventStream(
          "src/test/resources/data/opennlp/maxent/real-valued-weights-training-data.txt"); 
      DataIndexer testDataIndexer = new OnePassRealValueDataIndexer(rvfes1,1);
      // when
      QNModel trainedModel = new QNTrainer(5, 700, true).trainModel(ITERATIONS, testDataIndexer);
     
      ByteArrayOutputStream modelBytes = new ByteArrayOutputStream();
      GenericModelWriter modelWriter = new GenericModelWriter(trainedModel,
View Full Code Here
38
39
40
41
42
43
44
45
46
47
48
49
50
  @Test
  public void testDomainDimensionSanity() throws IOException {
    // given
    RealValueFileEventStream rvfes1 = new RealValueFileEventStream(
        "src/test/resources/data/opennlp/maxent/real-valued-weights-training-data.txt", "UTF-8"); 
    DataIndexer testDataIndexer = new OnePassRealValueDataIndexer(rvfes1,1);
    NegLogLikelihood objectFunction = new NegLogLikelihood(testDataIndexer);
    // when
    int correctDomainDimension = testDataIndexer.getPredLabels().length
        * testDataIndexer.getOutcomeLabels().length;
    // then
    assertEquals(correctDomainDimension, objectFunction.getDimension());
  }
View Full Code Here
52
53
54
55
56
57
58
59
60
61
62
  @Test
  public void testInitialSanity() throws IOException {
    // given
    RealValueFileEventStream rvfes1 = new RealValueFileEventStream(
        "src/test/resources/data/opennlp/maxent/real-valued-weights-training-data.txt", "UTF-8"); 
    DataIndexer testDataIndexer = new OnePassRealValueDataIndexer(rvfes1,1);
    NegLogLikelihood objectFunction = new NegLogLikelihood(testDataIndexer);
    // when
    double[] initial = objectFunction.getInitialPoint();
    // then
    for (int i = 0; i < initial.length; i++) {
View Full Code Here
67
68
69
70
71
72
73
74
75
76
77
  @Test
  public void testGradientSanity() throws IOException {
    // given
    RealValueFileEventStream rvfes1 = new RealValueFileEventStream(
        "src/test/resources/data/opennlp/maxent/real-valued-weights-training-data.txt", "UTF-8"); 
    DataIndexer testDataIndexer = new OnePassRealValueDataIndexer(rvfes1,1);
    NegLogLikelihood objectFunction = new NegLogLikelihood(testDataIndexer);
    // when
    double[] initial = objectFunction.getInitialPoint();
    double[] gradientAtInitial = objectFunction.gradientAt(initial);
    // then
View Full Code Here
TOP

Related Classes of opennlp.tools.ml.model.DataIndexer

  • opennlp.tools.ml.AbstractEventTrainer
  • opennlp.tools.ml.maxent.quasinewton.NegLogLikelihoodTest
  • opennlp.tools.ml.maxent.quasinewton.QNTrainerTest
  • opennlp.tools.ml.perceptron.SimplePerceptronSequenceTrainer
    • Copyright © 2018 www.massapicom. All rights reserved.
    All source code are property of their respective owners. Java is a trademark of Sun Microsystems, Inc and owned by ORACLE Inc. Contact coftware#gmail.com.