Package opennlp.tools.util

Examples of opennlp.tools.util.TrainingParameters

496
497
498
499
500
501
502
503
504
505
506
507
508
509
510
511
512
513
514
 
  public static ParserModel train(String languageCode,
      ObjectStream<Parse> parseSamples, HeadRules rules, int iterations, int cut)
      throws IOException {
   
    TrainingParameters params = new TrainingParameters();
    params.put("dict", TrainingParameters.CUTOFF_PARAM, Integer.toString(cut));

    params.put("tagger", TrainingParameters.CUTOFF_PARAM, Integer.toString(cut));
    params.put("tagger", TrainingParameters.ITERATIONS_PARAM, Integer.toString(iterations));
    params.put("chunker", TrainingParameters.CUTOFF_PARAM, Integer.toString(cut));
    params.put("chunker", TrainingParameters.ITERATIONS_PARAM, Integer.toString(iterations));
    params.put("check", TrainingParameters.CUTOFF_PARAM, Integer.toString(cut));
    params.put("check", TrainingParameters.ITERATIONS_PARAM, Integer.toString(iterations));
    params.put("build", TrainingParameters.CUTOFF_PARAM, Integer.toString(cut));
    params.put("build", TrainingParameters.ITERATIONS_PARAM, Integer.toString(iterations));
   
    return train(languageCode, parseSamples, rules, params);
  }
View Full Code Here
300
301
302
303
304
305
306
307
308
309
310
311
312
313
314
315
316
317
318
319
320
321
322
323
324
325
326
327
328
329
330
331
332
333
334
335
336
337
 
  // its optional, passing null is allowed
  public static TrainingParameters loadTrainingParameters(String paramFile,
      boolean supportSequenceTraining) {
   
    TrainingParameters params = null;
   
    if (paramFile != null) {
     
      checkInputFile("Training Parameter", new File(paramFile));
     
      InputStream paramsIn = null;
      try {
        paramsIn = new FileInputStream(new File(paramFile));
       
        params = new opennlp.tools.util.TrainingParameters(paramsIn);
      } catch (IOException e) {
        throw new TerminateToolException(-1, "Error during parameters loading: " + e.getMessage(), e);
      }
      finally {
        try {
          if (paramsIn != null)
            paramsIn.close();
        } catch (IOException e) {
          //sorry that this can fail
        }
      }
     
      if (!TrainUtil.isValid(params.getSettings())) {
        throw new TerminateToolException(1, "Training parameters file '" + paramFile + "' is invalid!");
      }
     
      if (!supportSequenceTraining && TrainUtil.isSequenceTraining(params.getSettings())) {
        throw new TerminateToolException(1, "Sequence training is not supported!");
      }
    }
   
    return params;
View Full Code Here
303
304
305
306
307
308
309
310
311
312
313
314
315
316
    try(FileInputStream inStream = new FileInputStream(inFile)){
      ObjectStream<String> lineStream = new PlainTextByLineStream(inStream, charset);
      ObjectStream<SentenceSample> sampleStream = new SentenceSampleStream(lineStream);

      // Training Parameters
      TrainingParameters mlParams = new TrainingParameters();
      mlParams.put(TrainingParameters.ALGORITHM_PARAM, "MAXENT");
      mlParams.put(TrainingParameters.ITERATIONS_PARAM, Integer.toString(iters));
      mlParams.put(TrainingParameters.CUTOFF_PARAM, Integer.toString(cut));

      // Abbreviations dictionary
      // TODO: Actually import a Dictionary of abbreviations
      Dictionary dict = new Dictionary();
View Full Code Here
39
40
41
42
43
44
45
46
47
48
49
50
51
  public void testSentenceDetector() throws IOException {

    InputStream in = getClass().getResourceAsStream(
        "/opennlp/tools/sentdetect/Sentences.txt");

    TrainingParameters mlParams = new TrainingParameters();
    mlParams.put(TrainingParameters.ITERATIONS_PARAM, Integer.toString(100));
    mlParams.put(TrainingParameters.CUTOFF_PARAM, Integer.toString(0));

    SentenceModel sentdetectModel = SentenceDetectorME.train(
        "en", new SentenceSampleStream(new PlainTextByLineStream(new InputStreamReader(in))), true, null, mlParams);

    assertEquals("en", sentdetectModel.getLanguage());
View Full Code Here
41
42
43
44
45
46
47
48
49
50
51
52
53
      new DocumentSample("0", new String[]{"x", "y", "z"}),
      new DocumentSample("0", new String[]{"x", "y", "z", "5", "6"}),
      new DocumentSample("0", new String[]{"x", "y", "z", "7", "8"})
    });

    TrainingParameters params = new TrainingParameters();
    params.put(TrainingParameters.ITERATIONS_PARAM, Integer.toString(100));
    params.put(TrainingParameters.CUTOFF_PARAM, Integer.toString(0));

    DoccatModel model = DocumentCategorizerME.train("x-unspecified", samples,
            params, new BagOfWordsFeatureGenerator());

    DocumentCategorizer doccat = new DocumentCategorizerME(model);
View Full Code Here
51
52
53
54
55
56
57
58
59
60
61
62
        new Span(0, 3)}));
    samples.add(new TokenSample("yes,", new Span[]{
        new Span(0, 3),
        new Span(3, 4)}));

    TrainingParameters mlParams = new TrainingParameters();
    mlParams.put(TrainingParameters.ITERATIONS_PARAM, Integer.toString(100));
    mlParams.put(TrainingParameters.CUTOFF_PARAM, Integer.toString(0));

    return TokenizerME.train("en", new CollectionObjectStream<TokenSample>(samples), true,
        mlParams);
  }
View Full Code Here
67
68
69
70
71
72
73
74
75
76
77
        "/opennlp/tools/tokenize/token.train");

    ObjectStream<TokenSample> samples = new TokenSampleStream(
        new PlainTextByLineStream(new InputStreamReader(trainDataIn, "UTF-8")));

    TrainingParameters mlParams = new TrainingParameters();
    mlParams.put(TrainingParameters.ITERATIONS_PARAM, Integer.toString(100));
    mlParams.put(TrainingParameters.CUTOFF_PARAM, Integer.toString(0));

    return TokenizerME.train("en", samples, true, mlParams);
  }
View Full Code Here
78
79
80
81
82
83
84
85
86
87
88
89
90
    String encoding = "UTF-8";

    ObjectStream<ChunkSample> sampleStream = new ChunkSampleStream(
        new PlainTextByLineStream(new InputStreamReader(in, encoding)));

    TrainingParameters params = new TrainingParameters();
    params.put(TrainingParameters.ITERATIONS_PARAM, Integer.toString(70));
    params.put(TrainingParameters.CUTOFF_PARAM, Integer.toString(1));

    ChunkerModel chunkerModel = ChunkerME.train("en", sampleStream, params, new ChunkerFactory());

    this.chunker = new ChunkerME(chunkerModel);
  }
View Full Code Here
491
492
493
494
495
496
497
498
499
500
501
502
503
504
505
506
507
508
509
  public static ParserModel train(String languageCode,
      ObjectStream<Parse> parseSamples, HeadRules rules, int iterations, int cut)
      throws IOException {

    TrainingParameters params = new TrainingParameters();
    params.put("dict", TrainingParameters.CUTOFF_PARAM, Integer.toString(cut));

    params.put("tagger", TrainingParameters.CUTOFF_PARAM, Integer.toString(cut));
    params.put("tagger", TrainingParameters.ITERATIONS_PARAM, Integer.toString(iterations));
    params.put("chunker", TrainingParameters.CUTOFF_PARAM, Integer.toString(cut));
    params.put("chunker", TrainingParameters.ITERATIONS_PARAM, Integer.toString(iterations));
    params.put("check", TrainingParameters.CUTOFF_PARAM, Integer.toString(cut));
    params.put("check", TrainingParameters.ITERATIONS_PARAM, Integer.toString(iterations));
    params.put("build", TrainingParameters.CUTOFF_PARAM, Integer.toString(cut));
    params.put("build", TrainingParameters.ITERATIONS_PARAM, Integer.toString(iterations));

    return train(languageCode, parseSamples, rules, params);
  }
View Full Code Here
607
608
609
610
611
612
613
614
615
616
   * @return A dictionary object.
   */
  public static Dictionary buildDictionary(ObjectStream<Parse> data, HeadRules rules, int cutoff)
      throws IOException {

    TrainingParameters params = new TrainingParameters();
    params.put("dict", TrainingParameters.CUTOFF_PARAM, Integer.toString(cutoff));

    return buildDictionary(data, rules, params);
  }
View Full Code Here
TOP

Related Classes of opennlp.tools.util.TrainingParameters

  • opennlp.tools.chunker.ChunkerMETest
  • opennlp.tools.cmdline.chunker.ChunkerCrossValidatorTool
  • opennlp.tools.cmdline.CmdLineUtil
  • opennlp.tools.cmdline.namefind.TokenNameFinderCrossValidatorTool
  • opennlp.tools.cmdline.postag.POSTaggerCrossValidatorTool
  • opennlp.tools.cmdline.sentdetect.SentenceDetectorCrossValidatorTool
  • opennlp.tools.cmdline.tokenizer.TokenizerTrainerTool
  • opennlp.tools.doccat.DocumentCategorizerMETest
  • opennlp.tools.ml.TrainerFactoryTest
  • opennlp.tools.namefind.NameFinderMETest
    • 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.