Examples of TFIDF

Examples of edu.udo.cs.wvtool.generic.vectorcreation.TFIDF

54
55
56
57
58
59
60
61
62
63
64
        FileWriter outFile = new FileWriter("wv.txt");
        WordVectorWriter wvw = new WordVectorWriter(outFile, true);

        config.setConfigurationRule(WVTConfiguration.STEP_OUTPUT, new WVTConfigurationFact(wvw));

        config.setConfigurationRule(WVTConfiguration.STEP_VECTOR_CREATION, new WVTConfigurationFact(new TFIDF()));

        // Create the vectors
        wvt.createVectors(list, config, wordList);

        // Alternatively: create word list and vectors together
View Full Code Here

Examples of edu.udo.cs.wvtool.generic.vectorcreation.TFIDF

77
78
79
80
81
82
83
84
85
86
87
        FileWriter outFile = new FileWriter("wv.txt");
        WordVectorWriter wvw = new WordVectorWriter(outFile, true);

        config.setConfigurationRule(WVTConfiguration.STEP_OUTPUT, new WVTConfigurationFact(wvw));

        config.setConfigurationRule(WVTConfiguration.STEP_VECTOR_CREATION, new WVTConfigurationFact(new TFIDF()));

        // Create the vectors
        wvt.createVectors(list, config, wordList);

        // Alternatively: create word list and vectors together
View Full Code Here

Examples of org.apache.mahout.utils.vectors.TFIDF

129
130
131
132
133
134
135
136
137
138
139
      doc.add(text);
      writer.addDocument(doc);
    }
    writer.close();
    IndexReader reader = IndexReader.open(directory, true);
    Weight weight = new TFIDF();
    TermInfo termInfo = new CachedTermInfo(reader, "content", 1, 100);
    VectorMapper mapper = new TFDFMapper(reader, weight, termInfo);
    LuceneIterable iterable = new LuceneIterable(reader, "id", "content", mapper);
   
    int i = 0;
View Full Code Here

Examples of org.apache.mahout.utils.vectors.TFIDF

62
63
64
65
66
67
68
69
70
71
72
    writer.close();
  }
 
  public void testIterable() throws Exception {
    IndexReader reader = IndexReader.open(directory, true);
    Weight weight = new TFIDF();
    TermInfo termInfo = new CachedTermInfo(reader, "content", 1, 100);
    VectorMapper mapper = new TFDFMapper(reader, weight, termInfo);
    LuceneIterable iterable = new LuceneIterable(reader, "id", "content", mapper);
   
    //TODO: do something more meaningful here
View Full Code Here

Examples of org.apache.mahout.utils.vectors.TFIDF

146
147
148
149
150
151
152
153
154
155
156
157
158
159
160
161
          if (cmdLine.hasOption(weightOpt)) {
            String wString = cmdLine.getValue(weightOpt).toString();
            if (wString.equalsIgnoreCase("tf")) {
              weight = new TF();
            } else if (wString.equalsIgnoreCase("tfidf")) {
              weight = new TFIDF();
            } else {
              throw new OptionException(weightOpt);
            }
          } else {
            weight = new TFIDF();
          }
          String field = cmdLine.getValue(fieldOpt).toString();
          int minDf = 1;
          if (cmdLine.hasOption(minDFOpt)) {
            minDf = Integer.parseInt(cmdLine.getValue(minDFOpt).toString());
View Full Code Here

Examples of org.apache.mahout.utils.vectors.TFIDF

60
61
62
63
64
65
66
67
68
69
70
    writer.close();
  }

  public void testIterable() throws Exception {
    IndexReader reader = IndexReader.open(directory, true);
    Weight weight = new TFIDF();
    TermInfo termInfo = new CachedTermInfo(reader, "content", 1, 100);
    VectorMapper mapper = new TFDFMapper(reader, weight, termInfo);
    LuceneIterable iterable = new LuceneIterable(reader, "id", "content", mapper);

    //TODO: do something more meaningful here
View Full Code Here

Examples of org.apache.mahout.utils.vectors.TFIDF

142
143
144
145
146
147
148
149
150
151
152
153
154
155
156
157
          if (cmdLine.hasOption(weightOpt)) {
            String wString = cmdLine.getValue(weightOpt).toString();
            if (wString.equalsIgnoreCase("tf")) {
              weight = new TF();
            } else if (wString.equalsIgnoreCase("tfidf")) {
              weight = new TFIDF();
            } else {
              throw new OptionException(weightOpt);
            }
          } else {
            weight = new TFIDF();
          }
          String field = cmdLine.getValue(fieldOpt).toString();
          int minDf = 1;
          if (cmdLine.hasOption(minDFOpt)) {
            minDf = Integer.parseInt(cmdLine.getValue(minDFOpt).toString());
View Full Code Here

Examples of org.apache.mahout.vectorizer.TFIDF

113
114
115
116
117
118
119
120
121
122
123
      doc.add(text);
      writer.addDocument(doc);
    }
    writer.close();
    IndexReader reader = IndexReader.open(directory, true);
    Weight weight = new TFIDF();
    TermInfo termInfo = new CachedTermInfo(reader, "content", 1, 100);
    VectorMapper mapper = new TFDFMapper(reader, weight, termInfo);
    Iterable<Vector> iterable = new LuceneIterable(reader, "id", "content", mapper);

    int i = 0;
View Full Code Here

Examples of org.apache.mahout.vectorizer.TFIDF

112
113
114
115
116
117
118
119
120
121
122
      doc.add(text);
      writer.addDocument(doc);
    }
    writer.close();
    IndexReader reader = IndexReader.open(directory, true);
    Weight weight = new TFIDF();
    TermInfo termInfo = new CachedTermInfo(reader, "content", 1, 100);

    int numTerms = 0;
    for (Iterator<TermEntry> it = termInfo.getAllEntries(); it.hasNext();) {
      it.next();
View Full Code Here

Examples of org.apache.mahout.vectorizer.TFIDF

67
68
69
70
71
72
73
74
75
76
77
  }

  @Test
  public void testIterable() throws Exception {
    IndexReader reader = IndexReader.open(directory, true);
    Weight weight = new TFIDF();
    TermInfo termInfo = new CachedTermInfo(reader, "content", 1, 100);
    VectorMapper mapper = new TFDFMapper(reader, weight, termInfo);
    LuceneIterable iterable = new LuceneIterable(reader, "id", "content", mapper);

    //TODO: do something more meaningful here
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.