Examples of KLDClassifier

Examples of com.github.pmerienne.trident.ml.nlp.KLDClassifier

41
42
43
44
45
46
47
48
49
50
51
52
53
54
  @Test
  public void testWithSmallWiki() {
    EnglishTokenizer tokenizer = new EnglishTokenizer();

    KLDClassifier kldClassifier = new KLDClassifier(2);
    kldClassifier.update(0, tokenizer.tokenize(NOSQL_WIKI));
    kldClassifier.update(0, tokenizer.tokenize(MYSQL_WIKI));
    kldClassifier.update(1, tokenizer.tokenize(LILIUM_WIKI));
    kldClassifier.update(1, tokenizer.tokenize(ROSE_WIKI));

    assertEquals(0, (int) kldClassifier.classify(tokenizer.tokenize(DATABASE_WIKI)));
    assertEquals(1, (int) kldClassifier.classify(tokenizer.tokenize(FLOWER_WIKI)));
  }
View Full Code Here

Examples of com.github.pmerienne.trident.ml.nlp.KLDClassifier

57
58
59
60
61
62
63
64
65
66
67
68
69
70
71
72
73
74
75
76
77
78
  public void testWithReuters() {
    List<TextInstance<Integer>> training = DatasetUtils.getTrainingFolds(0, 10, Datasets.getReutersSamples());
    List<TextInstance<Integer>> eval = DatasetUtils.getEvalFold(0, 10, Datasets.getReutersSamples());

    // Use 500 max words per class to speed up test
    KLDClassifier kldClassifier = new KLDClassifier(9, 500);

    // Train
    for (TextInstance<Integer> instance : training) {
      kldClassifier.update(instance.label, instance.tokens);
    }

    // Eval
    double evalSize = 0.0;
    double errorCount = 0.0;
    for (TextInstance<Integer> instance : eval) {
      int actual = kldClassifier.classify(instance.tokens);
      if (actual != instance.label) {
        errorCount++;
      }
      evalSize++;
    }
View Full Code Here

Examples of com.github.pmerienne.trident.ml.nlp.KLDClassifier

49
50
51
52
53
54
55
56
57
58
59
      TridentState classifierState = toppology.newStream("reutersData", new ReutersBatchSpout())
      // Transform raw data to text instance
          .each(new Fields("label", "text"), new TextInstanceCreator<Integer>(), new Fields("instance"))

          // Update text classifier
          .partitionPersist(new MemoryMapState.Factory(), new Fields("instance"), new TextClassifierUpdater<Integer>("newsClassifier", new KLDClassifier(9)));

      // Classification stream
      toppology.newDRPCStream("classify", localDRPC)
      // Convert DRPC args to text instance
          .each(new Fields("args"), new TextInstanceCreator<Integer>(false), new Fields("instance"))
View Full Code Here
TOP
Copyright © 2018 www.massapi.com. 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.