Package ivory.core.data.document

Examples of ivory.core.data.document.WeightedIntDocVector

164
165
166
167
168
169
170
171
172
173
174
    Configuration conf = IntegrationUtils.getBespinConfiguration();
    FileSystem fs = FileSystem.get(conf);

    SequenceFile.Reader reader;
    IntWritable key = new IntWritable();
    WeightedIntDocVector value = new WeightedIntDocVector();

    reader = new SequenceFile.Reader(fs.getConf(),
        SequenceFile.Reader.file(new Path(enwikiEn + "/test_wt-int-doc-vectors/part-00000")));
    reader.next(key, value);
    verifyIntDocVector(enIntDocVector1, value);
View Full Code Here
277
278
279
280
281
282
283
284
285
286
287
    Configuration conf = IntegrationUtils.getBespinConfiguration();
    FileSystem fs = FileSystem.get(conf);

    SequenceFile.Reader reader;
    IntWritable key = new IntWritable();
    WeightedIntDocVector value = new WeightedIntDocVector();

    reader = new SequenceFile.Reader(fs.getConf(),
        SequenceFile.Reader.file(new Path(dewikiEn + "/test_wt-int-doc-vectors/part-00000")));
    reader.next(key, value);
    verifyIntDocVector(deIntDocVector1, value);
View Full Code Here
161
162
163
164
165
166
167
168
169
170
171
    Configuration conf = IntegrationUtils.getBespinConfiguration();
    FileSystem fs = FileSystem.get(conf);

    SequenceFile.Reader reader;
    IntWritable key = new IntWritable();
    WeightedIntDocVector value = new WeightedIntDocVector();

    reader = new SequenceFile.Reader(fs.getConf(),
        SequenceFile.Reader.file(new Path(galagoIndex + "/test_wt-int-doc-vectors/part-00000")));
    reader.next(key, value);
    System.out.println("galagoInt1\n"+key+";"+value);
View Full Code Here
266
267
268
269
270
271
272
273
274
275
276
    Configuration conf = IntegrationUtils.getBespinConfiguration();
    FileSystem fs = FileSystem.get(conf);

    SequenceFile.Reader reader;
    IntWritable key = new IntWritable();
    WeightedIntDocVector value = new WeightedIntDocVector();

    reader = new SequenceFile.Reader(fs.getConf(),
        SequenceFile.Reader.file(new Path(opennlpIndex + "/test_wt-int-doc-vectors/part-00000")));
    reader.next(key, value);
    System.out.println("opennlpInt1\n"+key+";"+value);
View Full Code Here
265
266
267
268
269
270
271
272
273
274
275
276
277
278
279
280
281
282
283
284
285
286
287
288
289
290
    FileSystem fs = FileSystem.get(conf);

    SequenceFile.Reader reader;
    IntWritable key = new IntWritable();
    HMapIFW map = new HMapIFW();
    WeightedIntDocVector value = new WeightedIntDocVector();

    reader = new SequenceFile.Reader(fs.getConf(),
        SequenceFile.Reader.file(new Path(enwikiEn + "/test_wt-int-doc-vectors/part-00000")));

    reader.next(key, value);
    System.out.println("*** top 10 terms ***");
    map = value.getWeightedTerms();
    for ( MapIF.Entry entry : map.getEntriesSortedByValue(10)) {
      System.out.println(entry.getKey() + ": " + entry.getValue());
    }
    verifyIntDocVector(enIntDocVector1, value);

    reader.next(key, value);
    System.out.println("*** top 10 terms ***");
    map = value.getWeightedTerms();
    for ( MapIF.Entry entry : map.getEntriesSortedByValue(10)) {
      System.out.println(entry.getKey() + ": " + entry.getValue());
    }
    verifyIntDocVector(enIntDocVector2, value);
    reader.close();
View Full Code Here
399
400
401
402
403
404
405
406
407
408
409
410
411
412
413
414
415
416
417
418
419
420
421
422
423
    FileSystem fs = FileSystem.get(conf);

    SequenceFile.Reader reader;
    IntWritable key = new IntWritable();
    HMapIFW map = new HMapIFW();
    WeightedIntDocVector value = new WeightedIntDocVector();

    reader = new SequenceFile.Reader(fs.getConf(),
        SequenceFile.Reader.file(new Path(dewikiEn + "/test_wt-int-doc-vectors/part-00000")));
    reader.next(key, value);
    System.out.println("*** top 10 terms ***");
    map = value.getWeightedTerms();
    for ( MapIF.Entry entry : map.getEntriesSortedByValue(10)) {
      System.out.println(entry.getKey() + ": " + entry.getValue());
    }
    verifyIntDocVector(deIntDocVector1, value);

    reader.next(key, value);
    System.out.println("*** top 10 terms ***");
    map = value.getWeightedTerms();
    for ( MapIF.Entry entry : map.getEntriesSortedByValue(10)) {
      System.out.println(entry.getKey() + ": " + entry.getValue());
    }
    verifyIntDocVector(deIntDocVector2, value);
    reader.close();
View Full Code Here
262
263
264
265
266
267
268
269
270
271
272
273
274
275
276
277
278
279
280
281
282
283
284
285
286
287
    FileSystem fs = FileSystem.get(conf);

    SequenceFile.Reader reader;
    IntWritable key = new IntWritable();
    HMapIFW map = new HMapIFW();
    WeightedIntDocVector value = new WeightedIntDocVector();

    reader = new SequenceFile.Reader(fs.getConf(),
        SequenceFile.Reader.file(new Path(galagoIndex + "/test_wt-int-doc-vectors/part-00000")));

    reader.next(key, value);
    System.out.println("*** top 10 terms ***");
    map = value.getWeightedTerms();
    for ( MapIF.Entry entry : map.getEntriesSortedByValue(10)) {
      System.out.println(entry.getKey() + ": " + entry.getValue());
    }
    verifyIntDocVector(galagoIntDocVector1, value);

    reader.next(key, value);
    System.out.println("*** top 10 terms ***");
    map = value.getWeightedTerms();
    for ( MapIF.Entry entry : map.getEntriesSortedByValue(10)) {
      System.out.println(entry.getKey() + ": " + entry.getValue());
    }
    verifyIntDocVector(galagoIntDocVector2, value);
    reader.close();
View Full Code Here
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
    Configuration conf = IntegrationUtils.getBespinConfiguration();
    FileSystem fs = FileSystem.get(conf);

    SequenceFile.Reader reader;
    IntWritable key = new IntWritable();
    WeightedIntDocVector value = new WeightedIntDocVector();
    HMapIFW map = new HMapIFW();

    reader = new SequenceFile.Reader(fs.getConf(),
        SequenceFile.Reader.file(new Path(opennlpIndex + "/test_wt-int-doc-vectors/part-00000")));
    reader.next(key, value);
    map = value.getWeightedTerms();
    System.out.println("*** top 10 terms ***");
    for ( MapIF.Entry entry : map.getEntriesSortedByValue(10)) {
      System.out.println(entry.getKey() + ": " + entry.getValue());
    }
    verifyIntDocVector(opennlpIntDocVector1, value);

    reader.next(key, value);
    map = value.getWeightedTerms();
    System.out.println("*** top 10 terms ***");
    for ( MapIF.Entry entry : map.getEntriesSortedByValue(10)) {
      System.out.println(entry.getKey() + ": " + entry.getValue());
    }
    verifyIntDocVector(opennlpIntDocVector2, value);
View Full Code Here
140
141
142
143
144
145
146
147
148
149
150
  public void verifyIntDocVectorsEn() throws Exception {
    System.out.println("verifyIntDocVectorsEn");
    Configuration conf = IntegrationUtils.getBespinConfiguration();
    FileSystem fs = FileSystem.get(conf);

    WeightedIntDocVector value = findIntDocVector(enIntDocVector1Id, 2, fs, enwikiEn + "/wt-int-doc-vectors");
    System.out.println("enIntDocVector1\n"+value);
    verifyIntDocVector(enIntDocVector1, value);

    value = findIntDocVector(enIntDocVector2Id, 11, fs, enwikiEn + "/wt-int-doc-vectors");
    System.out.println("enIntDocVector2\n"+value);
View Full Code Here
217
218
219
220
221
222
223
224
225
226
227
  public void verifyIntDocVectorsDe() throws Exception {
    System.out.println("verifyIntDocVectorsDe");
    Configuration conf = IntegrationUtils.getBespinConfiguration();
    FileSystem fs = FileSystem.get(conf);

    WeightedIntDocVector value = findIntDocVector(deIntDocVector1Id, 2, fs, dewikiEn + "/wt-int-doc-vectors");
    System.out.println("deIntDocVector1\n"+value);
    verifyIntDocVector(deIntDocVector1, value);

    value = findIntDocVector(deIntDocVector2Id, 11, fs, dewikiEn + "/wt-int-doc-vectors");
    System.out.println("deIntDocVector2\n"+value);
View Full Code Here
TOP

Related Classes of ivory.core.data.document.WeightedIntDocVector

  • edu.umd.cloud9.io.map.HMapIFW
  • ivory.core.preprocess.BuildWeightedIntDocVectors$MyMapper
  • ivory.integration.wikipedia.VerifyWikipediaProcessingCrosslingual
  • ivory.integration.wikipedia.VerifyWikipediaProcessingMonolingual
  • ivory.lsh.eval.BruteForcePwsim$MyMapperDocVectors
  • ivory.lsh.eval.BruteForcePwsim$MyMapperWeightedIntDocVectors
    • 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.