Package org.apache.mahout.utils.vectors

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
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
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
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
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
TOP

Related Classes of org.apache.mahout.utils.vectors.TFIDF

  • org.apache.mahout.clustering.dirichlet.TestL1ModelClustering
  • org.apache.mahout.utils.vectors.lucene.Driver
  • org.apache.mahout.utils.vectors.lucene.LuceneIterableTest
    • 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.