Package opennlp.maxent

Examples of opennlp.maxent.GISModel

105
106
107
108
109
110
111
112
113
114
115
    Parser parser = new ChunkParser(chunker, tagger);
    AnswerTypeContextGenerator actg = new AnswerTypeContextGenerator(new File(wordnetDir));
    //<start id="atc.train"/>
    AnswerTypeEventStream es = new AnswerTypeEventStream(trainFile,
            actg, parser);
    GISModel model = GIS.trainModel(100, new TwoPassDataIndexer(es, 3));//<co id="atc.train.do"/>
    new DoccatModel("en", model).serialize(new FileOutputStream(outFile));
    /*
    <calloutlist>
        <callout arearefs="atc.train.do"><para>Using the event stream, which feeds us training examples, do the actual training using OpenNLP's Maxent classifier.</para></callout>
    </calloutlist>
View Full Code Here
72
73
74
75
76
77
78
79
    String[] outcomeLabels = getOutcomes();
    int[][] outcomePatterns = getOutcomePatterns();
    String[] predLabels = getPredicates();
    Context[] params = getParameters(outcomePatterns);

    return new GISModel(params, predLabels, outcomeLabels, correctionConstant,
        correctionParam);
  }
View Full Code Here
66
67
68
69
70
71
72
73
74
75
76
   * during IO operations on a temp file which is created during training occur.
   */
  public static POSModel train(String languageCode, ObjectStream<POSSample> samples, POSDictionary tagDictionary,
      Dictionary ngramDictionary, int cutoff, int iterations) throws IOException {

    GISModel posModel = opennlp.maxent.GIS.trainModel(iterations,
        new TwoPassDataIndexer(new POSSampleEventStream(samples,
        new DefaultPOSContextGenerator(ngramDictionary)), cutoff));

    return new POSModel(languageCode, posModel, tagDictionary, ngramDictionary);
  }
View Full Code Here
180
181
182
183
184
185
186
187
188
189
190
          }
        }
      }
      File inFile = new File(args[ai++]);
      File outFile = new File(args[ai++]);
      GISModel mod;

      EventStream es = new NameFinderEventStream(new PlainTextByLineDataStream(new InputStreamReader(new FileInputStream(inFile),encoding)));
      if (args.length > ai)
        mod = train(es, Integer.parseInt(args[ai++]), Integer.parseInt(args[ai++]));
      else
View Full Code Here
271
272
273
274
275
276
277
278
279
280
281
        }
      }
    }
    File inFile = new File(args[ai++]);
    File outFile = new File(args[ai++]);
    GISModel mod;
   
    try {
      EventStream es = new SDEventStream(new PlainTextByLineDataStream(new InputStreamReader(new FileInputStream(inFile),encoding)));

      if (args.length > ai)
View Full Code Here
316
317
318
319
320
321
322
323
324
325
326
      int iterations = 100;
      if (args.length > ai) {
        cutoff = Integer.parseInt(args[ai++]);
        iterations = Integer.parseInt(args[ai++]);
      }
      GISModel mod;
      if (dict != null) {
        System.err.println("Building dictionary");
        MutableDictionary mdict = new MutableDictionary(cutoff);
        DataStream data = new opennlp.maxent.PlainTextByLineDataStream(new java.io.FileReader(inFile));
        while(data.hasNext()) {
View Full Code Here
166
167
168
169
170
171
172
173
174
175
176
      System.err.println("space-delimited triple of \"word pos outcome\".  Sentence breaks are indicated by blank lines.");
      System.exit(1);
    }
    java.io.File inFile = new java.io.File(args[0]);
    java.io.File outFile = new java.io.File(args[1]);
    GISModel mod;
    opennlp.maxent.EventStream es = new ChunkerEventStream(new opennlp.maxent.PlainTextByLineDataStream(new java.io.FileReader(inFile)));
    if (args.length > 3)
      mod = train(es, Integer.parseInt(args[2]), Integer.parseInt(args[3]));
    else
      mod = train(es, 100, 5);
View Full Code Here
767
768
769
770
771
772
773
774
775
776
777
778
779
780
781
782
783
784
785
786
787
788
789
790
791
792
793
794
795
796
797
798
799
800
801
802
    if (tag || all) {
      System.err.println("Training tagger");
      System.err.println("Loading Dictionary");
      Dictionary tridict = new Dictionary(dictFile.toString());
      opennlp.maxent.EventStream tes = new ParserEventStream(new opennlp.maxent.PlainTextByLineDataStream(new java.io.FileReader(inFile)), rules, EventTypeEnum.TAG,tridict);
      GISModel tagModel = train(tes, iterations, cutoff);
      System.out.println("Saving the tagger model as: " + tagFile);
      new opennlp.maxent.io.SuffixSensitiveGISModelWriter(tagModel, tagFile).persist();
    }

    if (chunk || all) {
      System.err.println("Training chunker");
      opennlp.maxent.EventStream ces = new ParserEventStream(new opennlp.maxent.PlainTextByLineDataStream(new java.io.FileReader(inFile)), rules, EventTypeEnum.CHUNK);
      GISModel chunkModel = train(ces, iterations, cutoff);
      System.out.println("Saving the chunker model as: " + chunkFile);
      new opennlp.maxent.io.SuffixSensitiveGISModelWriter(chunkModel, chunkFile).persist();
    }

    if (build || all) {
      System.err.println("Loading Dictionary");
      Dictionary tridict = new Dictionary(dictFile.toString());
      System.err.println("Training builder");
      opennlp.maxent.EventStream bes = new ParserEventStream(new opennlp.maxent.PlainTextByLineDataStream(new java.io.FileReader(inFile)), rules, EventTypeEnum.BUILD,tridict);
      GISModel buildModel = train(bes, iterations, cutoff);
      System.out.println("Saving the build model as: " + buildFile);
      new opennlp.maxent.io.SuffixSensitiveGISModelWriter(buildModel, buildFile).persist();
    }

    if (check || all) {
      System.err.println("Training checker");
      opennlp.maxent.EventStream kes = new ParserEventStream(new opennlp.maxent.PlainTextByLineDataStream(new java.io.FileReader(inFile)), rules, EventTypeEnum.CHECK);
      GISModel checkModel = train(kes, iterations, cutoff);
      System.out.println("Saving the check model as: " + checkFile);
      new opennlp.maxent.io.SuffixSensitiveGISModelWriter(checkModel, checkFile).persist();
    }
  }
View Full Code Here
197
198
199
200
201
202
203
204
    }
    return (Span[]) tokens.toArray(new Span[tokens.size()]);
  }

  public static void train(EventStream evc, File output) throws IOException {
    GISModel tokModel = opennlp.maxent.GIS.trainModel(100,new TwoPassDataIndexer(evc, 5),true);
    new SuffixSensitiveGISModelWriter(tokModel, output).persist();
  }
View Full Code Here
189
190
191
192
193
194
195
196
197
198
199
200
201
   
    for(NamedEntity entity : entities) {
      List<String> features = fe.getFeatures(entity);
      String type = entity.getType();
      if(modelsByType.containsKey(type)) {
        GISModel model = modelsByType.get(type);
        if(model.getNumOutcomes() == 2) {
          double prob = model.eval(features.toArray(new String[0]))[model.getIndex("T")];
          entity.setConfidence(prob);
        }
      }
    }
  }
View Full Code Here
TOP

Related Classes of opennlp.maxent.GISModel

  • com.tamingtext.qa.AnswerTypeClassifier
  • edu.washington.cs.knowitall.extractor.conf.OpenNlpClassifierConfFunctionTest
  • edu.washington.cs.knowitall.extractor.conf.ReVerbOpenNlpClassifierTrainer
  • edu.washington.cs.knowitall.extractor.conf.ReVerbOpenNlpConfFunction
  • edu.washington.cs.knowitall.extractor.conf.WekaClassifierConfFunctionTest
  • opennlp.maxent.io.GISModelReader
  • opennlp.model.AbstractModel
  • opennlp.model.Context
  • opennlp.model.UniformPrior
  • opennlp.tools.chunker.ChunkerME
    • 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.