Package opennlp.tools.dictionary

Examples of opennlp.tools.dictionary.Dictionary$StringListWrapper

116
117
118
119
120
121
122
123
124
125
126
127
128
129
      System.exit(1);
    }

    int ai = 0;

    Dictionary abbreviations = null;
    if ("-abbreviationsDictionary".equals(args[ai])) {
      ai++;
      abbreviations = new Dictionary(new FileInputStream(args[ai++]));
    }

    boolean useTokenEnd = false;
    if ("-useTokenEnd".equals(args[ai])) {
      useTokenEnd = true;
View Full Code Here
84
85
86
87
88
89
90
91
92
93
94
    if (params.getMisclassified()) {
      listener = new TokenEvaluationErrorListener();
    }
   
    try {
      Dictionary dict = TokenizerTrainerTool.loadDict(params.getAbbDict());

      validator = new opennlp.tools.tokenize.TokenizerCrossValidator(
          params.getLang(), dict, params.getAlphaNumOpt(), mlParams, listener);

      validator.evaluate(sampleStream, params.getFolds());
View Full Code Here
283
284
285
286
287
288
289
290
291
292
293
  public static ParserModel train(String languageCode, ObjectStream<Parse> parseSamples, HeadRules rules, TrainingParameters mlParams)
          throws IOException {
   
    System.err.println("Building dictionary");
   
    Dictionary mdict = buildDictionary(parseSamples, rules, mlParams);
   
    parseSamples.reset();
   
    Map<String, String> manifestInfoEntries = new HashMap<String, String>();
   
View Full Code Here
112
113
114
115
116
117
118
119
120
121
122
    while (partitioner.hasNext()) {

      CrossValidationPartitioner.TrainingSampleStream<POSSample> trainingSampleStream = partitioner
          .next();
     
      Dictionary ngramDict = null;
      if (this.ngramDictionary == null) {
        if(this.ngramCutoff != null) {
          System.err.print("Building ngram dictionary ... ");
          try {
            ngramDict = POSTaggerME.buildNGramDictionary(trainingSampleStream,
View Full Code Here
71
72
73
74
75
76
77
78
79
80
81
82
    OutputStream out = null;
    try {
      in = new InputStreamReader(new FileInputStream(dictInFile), encoding);
      out = new FileOutputStream(dictOutFile);

      Dictionary dict = Dictionary.parseOneEntryPerLine(in);
      dict.serialize(out);

    } catch (IOException e) {
      CmdLineUtil.printTrainingIoError(e);
      throw new TerminateToolException(-1);
    } finally {
View Full Code Here
383
384
385
386
387
388
389
390
391
392
393
394
395
396
397
398
399
400
401
402
403
404
405
406
407
408
409
410
411
412
413
414
415
  public static Dictionary createOptionalDictionary(UimaContext context, String parameter)
    throws ResourceInitializationException {
  String dictionaryName = CasConsumerUtil.getOptionalStringParameter(
    context, parameter);
 
  Dictionary dictionary = null;

  if (dictionaryName != null) {

      Logger logger = context.getLogger();

      try {

    InputStream dictIn = CasConsumerUtil.getOptionalResourceAsStream(context,
      dictionaryName);

    if (dictIn == null) {
      String message = "The dictionary file " + dictionaryName +
      " does not exist!";

      if (logger.isLoggable(Level.WARNING)) {
          logger.log(Level.WARNING, message);
      }
     
      return null;
    }
   
    dictionary = new Dictionary(dictIn);

      } catch (IOException e) {
      // if this fails just print error message and continue
      String message = "IOException during dictionary reading, "
        + "running without dictionary: " + e.getMessage();
View Full Code Here
105
106
107
108
109
110
111
112
113
114
115
116
117
118
119
120
121
   *    built from the input file.
   * @throws IOException
   */
  public static Dictionary createDictionary(ObjectStream<StringList> sampleStream) throws IOException {

    Dictionary mNameDictionary = new Dictionary(true);
    StringList entry;

    entry = sampleStream.read();
    while (entry != null) {
      if (!mNameDictionary.contains(entry)) {
        mNameDictionary.put(entry);
      }
      entry = sampleStream.read();
    }

    return mNameDictionary;
View Full Code Here
145
146
147
148
149
150
151
152
153
154
155
156
157
158
159
160
161
162
163
164
165
166
167
168
169
170
171
172
173
174
175
176
    FileInputStream sampleDataIn = CmdLineUtil.openInFile(testData);
    ObjectStream<StringList> sampleStream = new NameFinderCensus90NameStream(sampleDataIn,
        Charset.forName(params.getEncoding()));
   
    Dictionary mDictionary;
    try {
      System.out.println("Creating Dictionary...");
      mDictionary = createDictionary(sampleStream);
    } catch (IOException e) {
      CmdLineUtil.printTrainingIoError(e);
      throw new TerminateToolException(-1);
    } finally {
      try {
        sampleStream.close();
      } catch(IOException e) {
        // sorry this can fail..
      }
    }

    System.out.println("Saving Dictionary...");
   
    OutputStream out = null;
   
    try {
      out = new FileOutputStream(dictOutFile);
      mDictionary.serialize(out);
    } catch (IOException ex) {
      System.err.println("Error during write to dictionary file: " + ex.getMessage());
      throw new TerminateToolException(-1);
    }
    finally {
View Full Code Here
51
52
53
54
55
56
57
58
59
60
61
62
63
64
65
66
67
68
69
70
   * Note: Do all initialization in this method, do not use the constructor.
   */
  public void initialize()
      throws ResourceInitializationException {
 
    Dictionary nameFinderDictionary;
   
    try {
      String modelName = AnnotatorUtil.getRequiredStringParameter(context,
          UimaUtil.DICTIONARY_PARAMETER);
     
      InputStream inModel = AnnotatorUtil
          .getResourceAsStream(context, modelName);
     
      nameFinderDictionary = new Dictionary(inModel);

    } catch (IOException e) {
      throw new ResourceInitializationException(
        ExceptionMessages.MESSAGE_CATALOG,
        ExceptionMessages.IO_ERROR_DICTIONARY_READING,
View Full Code Here
493
494
495
496
497
498
499
500
501
502
503
504
505
506
507
508
509
510
511
512
513
514
515
516
517
518
519
520
521
522
523
524
525
      String dictionaryParameter) throws ResourceInitializationException {

    String dictionaryName = AnnotatorUtil.getOptionalStringParameter(context,
        dictionaryParameter);

    Dictionary dictionary = null;

    if (dictionaryName != null) {

      Logger logger = context.getLogger();

      try {

        InputStream dictIn = AnnotatorUtil.getOptionalResourceAsStream(context,
            dictionaryName);

        if (dictIn == null) {
          String message = "The dictionary file " + dictionaryName
              + " does not exist!";

          if (logger.isLoggable(Level.WARNING)) {
            logger.log(Level.WARNING, message);
          }

          return null;
        }

        dictionary = new Dictionary(dictIn);

      } catch (IOException e) {
        // if this fails just print error message and continue
        String message = "IOException during dictionary reading, "
            + "running without dictionary: " + e.getMessage();
View Full Code Here
TOP

Related Classes of opennlp.tools.dictionary.Dictionary$StringListWrapper

  • io.lumify.opennlpDictionary.OpenNLPDictionaryExtractorGraphPropertyWorker
  • io.lumify.opennlpDictionary.OpenNLPDictionaryExtractorGraphPropertyWorkerTest
  • opennlp.tools.cmdline.dictionary.DictionaryBuilderTool
  • opennlp.tools.cmdline.namefind.CensusDictionaryCreatorTool
  • opennlp.tools.cmdline.parser.BuildModelUpdaterTool
  • opennlp.tools.cmdline.parser.CheckModelUpdaterTool
  • opennlp.tools.cmdline.parser.ParserTrainerTool
  • opennlp.tools.cmdline.postag.POSTaggerTrainerTool
  • opennlp.tools.cmdline.sentdetect.SentenceDetectorCrossValidatorTool
  • opennlp.tools.cmdline.sentdetect.SentenceDetectorTrainerTool
    • 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.