Package org.apache.mahout.clustering.dirichlet.models

Examples of org.apache.mahout.clustering.dirichlet.models.DistributionDescription

110
111
112
113
114
115
116
117
118
119
120
121
    boolean runClustering = hasOption(DefaultOptionCreator.CLUSTERING_OPTION);
    boolean runSequential =
        getOption(DefaultOptionCreator.METHOD_OPTION).equalsIgnoreCase(DefaultOptionCreator.SEQUENTIAL_METHOD);
    int prototypeSize = readPrototypeSize(input);

    DistributionDescription description =
        new DistributionDescription(modelFactory, modelPrototype, distanceMeasure, prototypeSize);

    run(getConf(),
        input,
        output,
        description,
View Full Code Here
245
246
247
248
249
250
251
252
253
254
255
256
257
258
259
260
261
262
263
264
265
266
267
268
269
270
271
272
273
274
275
    generateSamples(100, 0, 2, 0.3);
    generateSamples(100, 2, 2, 1);
    ClusteringTestUtils.writePointsToFile(sampleData, getTestTempFilePath("input/data.txt"), fs, conf);
    // Now run the driver using the run() method. Others can use runJob() as before
    Integer maxIterations = 5;
    DistributionDescription description =
        new DistributionDescription(GaussianClusterDistribution.class.getName(),
                                    DenseVector.class.getName(),
                                    null,
                                    2);
    String[] args = { optKey(DefaultOptionCreator.INPUT_OPTION), getTestTempDirPath("input").toString(),
        optKey(DefaultOptionCreator.OUTPUT_OPTION), getTestTempDirPath("output").toString(),
        optKey(DirichletDriver.MODEL_DISTRIBUTION_CLASS_OPTION), description.getModelFactory(),
        optKey(DirichletDriver.MODEL_PROTOTYPE_CLASS_OPTION), description.getModelPrototype(),
        optKey(DefaultOptionCreator.NUM_CLUSTERS_OPTION), "20", optKey(DefaultOptionCreator.MAX_ITERATIONS_OPTION),
        maxIterations.toString(), optKey(DirichletDriver.ALPHA_OPTION), "1.0",
        optKey(DefaultOptionCreator.OVERWRITE_OPTION),
        optKey(DefaultOptionCreator.CLUSTERING_OPTION), optKey(DefaultOptionCreator.METHOD_OPTION),
        DefaultOptionCreator.SEQUENTIAL_METHOD };
    DirichletDriver dirichletDriver = new DirichletDriver();
    dirichletDriver.setConf(conf);
    dirichletDriver.run(args);
    // and inspect results
    Collection<List<DirichletCluster>> clusters = Lists.newArrayList();
    Configuration conf = new Configuration();
    conf.set(DirichletDriver.MODEL_DISTRIBUTION_KEY, description.toString());
    conf.set(DirichletDriver.NUM_CLUSTERS_KEY, "20");
    conf.set(DirichletDriver.ALPHA_0_KEY, "1.0");
    for (int i = 0; i <= maxIterations; i++) {
      conf.set(DirichletDriver.STATE_IN_KEY, new Path(getTestTempDirPath("output"), "clusters-" + i).toString());
      clusters.add(DirichletMapper.getDirichletState(conf).getClusters());
View Full Code Here
285
286
287
288
289
290
291
292
293
294
295
296
297
298
299
300
301
302
303
304
305
306
307
308
309
310
311
312
    generateSamples(100, 0, 2, 0.3);
    generateSamples(100, 2, 2, 1);
    ClusteringTestUtils.writePointsToFile(sampleData, getTestTempFilePath("input/data.txt"), fs, conf);
    // Now run the driver using the run() method. Others can use runJob() as before
    Integer maxIterations = 5;
    DistributionDescription description =
        new DistributionDescription(GaussianClusterDistribution.class.getName(),
                                    DenseVector.class.getName(),
                                    null,
                                    2);
    String[] args = { optKey(DefaultOptionCreator.INPUT_OPTION), getTestTempDirPath("input").toString(),
        optKey(DefaultOptionCreator.OUTPUT_OPTION), getTestTempDirPath("output").toString(),
        optKey(DirichletDriver.MODEL_DISTRIBUTION_CLASS_OPTION), description.getModelFactory(),
        optKey(DirichletDriver.MODEL_PROTOTYPE_CLASS_OPTION), description.getModelPrototype(),
        optKey(DefaultOptionCreator.NUM_CLUSTERS_OPTION), "20", optKey(DefaultOptionCreator.MAX_ITERATIONS_OPTION),
        maxIterations.toString(), optKey(DirichletDriver.ALPHA_OPTION), "1.0",
        optKey(DefaultOptionCreator.OVERWRITE_OPTION),
        optKey(DefaultOptionCreator.CLUSTERING_OPTION)};
    ToolRunner.run(new Configuration(), new DirichletDriver(), args);
    // and inspect results
    Collection<List<DirichletCluster>> clusters = Lists.newArrayList();
    Configuration conf = new Configuration();
    conf.set(DirichletDriver.MODEL_DISTRIBUTION_KEY, description.toString());
    conf.set(DirichletDriver.NUM_CLUSTERS_KEY, "20");
    conf.set(DirichletDriver.ALPHA_0_KEY, "1.0");
    for (int i = 0; i <= maxIterations; i++) {
      conf.set(DirichletDriver.STATE_IN_KEY, new Path(getTestTempDirPath("output"), "clusters-" + i).toString());
      clusters.add(DirichletMapper.getDirichletState(conf).getClusters());
View Full Code Here
318
319
320
321
322
323
324
325
326
327
328
329
330
331
332
333
334
335
336
337
338
339
340
341
342
343
344
345
346
347
  @Test
  public void testDriverMnRIterations() throws Exception {
    generate4Datasets();
    // Now run the driver
    int maxIterations = 3;
    DistributionDescription description =
        new DistributionDescription(GaussianClusterDistribution.class.getName(),
                                    DenseVector.class.getName(),
                                    null,
                                    2);
    Configuration conf = new Configuration();
    DirichletDriver.run(conf,
                        getTestTempDirPath("input"),
                        getTestTempDirPath("output"),
                        description,
                        20,
                        maxIterations,
                        1.0,
                        false,
                        true,
                        0,
                        false);
    // and inspect results
    Collection<List<DirichletCluster>> clusters = Lists.newArrayList();
    conf.set(DirichletDriver.MODEL_DISTRIBUTION_KEY, description.toString());
    conf.set(DirichletDriver.NUM_CLUSTERS_KEY, "20");
    conf.set(DirichletDriver.ALPHA_0_KEY, "1.0");
    for (int i = 0; i <= maxIterations; i++) {
      conf.set(DirichletDriver.STATE_IN_KEY, new Path(getTestTempDirPath("output"), "clusters-" + i).toString());
      clusters.add(DirichletMapper.getDirichletState(conf).getClusters());
View Full Code Here
355
356
357
358
359
360
361
362
363
364
365
366
367
368
369
370
371
372
373
374
375
376
377
378
379
380
381
382
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
416
417
418
419
    generateAsymmetricSamples(100, 0, 0, 0.5, 3.0);
    generateAsymmetricSamples(100, 0, 3, 0.3, 4.0);
    ClusteringTestUtils.writePointsToFile(sampleData, getTestTempFilePath("input/data.txt"), fs, conf);
    // Now run the driver using the run() method. Others can use runJob() as before
    MahalanobisDistanceMeasure measure = new MahalanobisDistanceMeasure();
    DistributionDescription description =
        new DistributionDescription(DistanceMeasureClusterDistribution.class.getName(),
                                    DenseVector.class.getName(),
                                    MahalanobisDistanceMeasure.class.getName(),
                                    2);

    Vector meanVector = new DenseVector(new double[] { 0.0, 0.0 });
    measure.setMeanVector(meanVector);
    Matrix m= new DenseMatrix(new double [][] {{0.5, 0.0}, {0.0, 4.0}});
    measure.setCovarianceMatrix(m);

    Path inverseCovarianceFile =
        new Path(getTestTempDirPath("mahalanobis"), "MahalanobisDistanceMeasureInverseCovarianceFile");
    conf.set("MahalanobisDistanceMeasure.inverseCovarianceFile", inverseCovarianceFile.toString());
    FileSystem fs = FileSystem.get(inverseCovarianceFile.toUri(), conf);
    MatrixWritable inverseCovarianceMatrix = new MatrixWritable(measure.getInverseCovarianceMatrix());
    DataOutputStream out = fs.create(inverseCovarianceFile);
    try {
      inverseCovarianceMatrix.write(out);
    } finally {
      Closeables.closeQuietly(out);
    }

    Path meanVectorFile = new Path(getTestTempDirPath("mahalanobis"), "MahalanobisDistanceMeasureMeanVectorFile");
    conf.set("MahalanobisDistanceMeasure.meanVectorFile", meanVectorFile.toString());
    fs = FileSystem.get(meanVectorFile.toUri(), conf);
    VectorWritable meanVectorWritable = new VectorWritable(meanVector);
    out = fs.create(meanVectorFile);
    try {
      meanVectorWritable.write(out);
    } finally {
      Closeables.closeQuietly(out);
    }

    conf.set("MahalanobisDistanceMeasure.maxtrixClass", MatrixWritable.class.getName());
    conf.set("MahalanobisDistanceMeasure.vectorClass", VectorWritable.class.getName());

    Integer maxIterations = 5;
    String[] args = { optKey(DefaultOptionCreator.INPUT_OPTION), getTestTempDirPath("input").toString(),
        optKey(DefaultOptionCreator.OUTPUT_OPTION), getTestTempDirPath("output").toString(),
        optKey(DirichletDriver.MODEL_DISTRIBUTION_CLASS_OPTION), description.getModelFactory(),
        optKey(DefaultOptionCreator.DISTANCE_MEASURE_OPTION), description.getDistanceMeasure(),
        optKey(DirichletDriver.MODEL_PROTOTYPE_CLASS_OPTION), description.getModelPrototype(),
        optKey(DefaultOptionCreator.NUM_CLUSTERS_OPTION), "20", optKey(DefaultOptionCreator.MAX_ITERATIONS_OPTION),
        maxIterations.toString(), optKey(DirichletDriver.ALPHA_OPTION), "1.0",
        optKey(DefaultOptionCreator.OVERWRITE_OPTION),
        optKey(DefaultOptionCreator.CLUSTERING_OPTION), optKey(DefaultOptionCreator.METHOD_OPTION),
        DefaultOptionCreator.SEQUENTIAL_METHOD };
    DirichletDriver dirichletDriver = new DirichletDriver();
    dirichletDriver.setConf(conf);
    dirichletDriver.run(args);
    // and inspect results
    Collection<List<DirichletCluster>> clusters = Lists.newArrayList();
    Configuration conf = new Configuration();
    conf.set(DirichletDriver.MODEL_DISTRIBUTION_KEY, description.toString());
    conf.set(DirichletDriver.NUM_CLUSTERS_KEY, "20");
    conf.set(DirichletDriver.ALPHA_0_KEY, "1.0");
    for (int i = 0; i <= maxIterations; i++) {
      conf.set(DirichletDriver.STATE_IN_KEY, new Path(getTestTempDirPath("output"), "clusters-" + i).toString());
      clusters.add(DirichletMapper.getDirichletState(conf).getClusters());
View Full Code Here
428
429
430
431
432
433
434
435
436
437
438
439
440
441
442
443
444
445
446
447
448
449
450
451
452
453
454
455
456
457
458
459
460
461
462
463
464
465
466
467
468
469
470
471
472
473
474
475
476
477
478
479
480
481
482
483
484
485
486
487
488
489
490
491
    generateAsymmetricSamples(100, 0, 3, 0.3, 4.0);
    ClusteringTestUtils.writePointsToFile(sampleData, getTestTempFilePath("input/data.txt"), fs, conf);
    // Now run the driver using the run() method. Others can use runJob() as before

    MahalanobisDistanceMeasure measure = new MahalanobisDistanceMeasure();
    DistributionDescription description =
        new DistributionDescription(DistanceMeasureClusterDistribution.class.getName(),
                                    DenseVector.class.getName(),
                                    MahalanobisDistanceMeasure.class.getName(),
                                    2);

    Vector meanVector = new DenseVector(new double[]{0.0, 0.0});
    measure.setMeanVector(meanVector);
    Matrix m = new DenseMatrix(new double [][] {{0.5, 0.0}, {0.0, 4.0}});
    measure.setCovarianceMatrix(m);

    Path inverseCovarianceFile =
        new Path(getTestTempDirPath("mahalanobis"), "MahalanobisDistanceMeasureInverseCovarianceFile");
    conf.set("MahalanobisDistanceMeasure.inverseCovarianceFile", inverseCovarianceFile.toString());
    FileSystem fs = FileSystem.get(inverseCovarianceFile.toUri(), conf);
    MatrixWritable inverseCovarianceMatrix = new MatrixWritable(measure.getInverseCovarianceMatrix());
    DataOutputStream out = fs.create(inverseCovarianceFile);
    try {
      inverseCovarianceMatrix.write(out);
    } finally {
      Closeables.closeQuietly(out);
    }

    Path meanVectorFile = new Path(getTestTempDirPath("mahalanobis"), "MahalanobisDistanceMeasureMeanVectorFile");
    conf.set("MahalanobisDistanceMeasure.meanVectorFile", meanVectorFile.toString());
    fs = FileSystem.get(meanVectorFile.toUri(), conf);
    VectorWritable meanVectorWritable = new VectorWritable(meanVector);
    out = fs.create(meanVectorFile);
    try {
      meanVectorWritable.write(out);
    } finally {
      Closeables.closeQuietly(out);
    }

    conf.set("MahalanobisDistanceMeasure.maxtrixClass", MatrixWritable.class.getName());
    conf.set("MahalanobisDistanceMeasure.vectorClass", VectorWritable.class.getName());

    Integer maxIterations = 5;
    String[] args = { optKey(DefaultOptionCreator.INPUT_OPTION), getTestTempDirPath("input").toString(),
        optKey(DefaultOptionCreator.OUTPUT_OPTION), getTestTempDirPath("output").toString(),
        optKey(DirichletDriver.MODEL_DISTRIBUTION_CLASS_OPTION), description.getModelFactory(),
        optKey(DefaultOptionCreator.DISTANCE_MEASURE_OPTION), description.getDistanceMeasure(),
        optKey(DirichletDriver.MODEL_PROTOTYPE_CLASS_OPTION), description.getModelPrototype(),
        optKey(DefaultOptionCreator.NUM_CLUSTERS_OPTION), "20", optKey(DefaultOptionCreator.MAX_ITERATIONS_OPTION),
        maxIterations.toString(), optKey(DirichletDriver.ALPHA_OPTION), "1.0",
        optKey(DefaultOptionCreator.OVERWRITE_OPTION),
        optKey(DefaultOptionCreator.CLUSTERING_OPTION)};
    Tool dirichletDriver = new DirichletDriver();
    dirichletDriver.setConf(conf);
    ToolRunner.run(conf, dirichletDriver, args);
    // and inspect results
    Collection<List<DirichletCluster>> clusters = Lists.newArrayList();
    Configuration conf = new Configuration();
    conf.set(DirichletDriver.MODEL_DISTRIBUTION_KEY, description.toString());
    conf.set(DirichletDriver.NUM_CLUSTERS_KEY, "20");
    conf.set(DirichletDriver.ALPHA_0_KEY, "1.0");
    for (int i = 0; i <= maxIterations; i++) {
      conf.set(DirichletDriver.STATE_IN_KEY, new Path(getTestTempDirPath("output"), "clusters-" + i).toString());
      clusters.add(DirichletMapper.getDirichletState(conf).getClusters());
View Full Code Here
248
249
250
251
252
253
254
255
256
257
258
 
  @Test
  public void testDirichlet2() throws Exception {
    Path output = getTestTempDirPath("output");
    NamedVector prototype = (NamedVector) sampleData.get(0).get();
    DistributionDescription description = new DistributionDescription(
        GaussianClusterDistribution.class.getName(),
        RandomAccessSparseVector.class.getName(), null, prototype.getDelegate()
            .size());
    Configuration conf = new Configuration();
    DirichletDriver.run(conf, getTestTempDirPath("testdata"), output,
View Full Code Here
265
266
267
268
269
270
271
272
273
274
275
 
  @Test
  public void testDirichlet3() throws Exception {
    Path output = getTestTempDirPath("output");
    NamedVector prototype = (NamedVector) sampleData.get(0).get();
    DistributionDescription description = new DistributionDescription(
        DistanceMeasureClusterDistribution.class.getName(),
        RandomAccessSparseVector.class.getName(),
        ManhattanDistanceMeasure.class.getName(), prototype.getDelegate()
            .size());
    Configuration conf = new Configuration();
View Full Code Here
51
52
53
54
55
56
57
58
59
60
61
62
      ToolRunner.run(new Configuration(), new Job(), args);
    } else {
      log.info("Running with default arguments");
      Path output = new Path("output");
      HadoopUtil.delete(new Configuration(), output);
      DistributionDescription description =
          new DistributionDescription(GaussianClusterDistribution.class.getName(),
                                      RandomAccessSparseVector.class.getName(),
                                      null,
                                      60);
      run(new Path("testdata"), output, description, 10, 5, 1.0, true, 0);
    }
View Full Code Here
103
104
105
106
107
108
109
110
111
112
113
    int numModels = Integer.parseInt(getOption(DefaultOptionCreator.NUM_CLUSTERS_OPTION));
    int maxIterations = Integer.parseInt(getOption(DefaultOptionCreator.MAX_ITERATIONS_OPTION));
    boolean emitMostLikely = Boolean.parseBoolean(getOption(DefaultOptionCreator.EMIT_MOST_LIKELY_OPTION));
    double threshold = Double.parseDouble(getOption(DefaultOptionCreator.THRESHOLD_OPTION));
    double alpha0 = Double.parseDouble(getOption(DirichletDriver.ALPHA_OPTION));
    DistributionDescription description =
        new DistributionDescription(modelFactory, modelPrototype, distanceMeasure, 60);

    run(input, output, description, numModels, maxIterations, alpha0, emitMostLikely, threshold);
    return 0;
  }
View Full Code Here
TOP

Related Classes of org.apache.mahout.clustering.dirichlet.models.DistributionDescription

  • org.apache.mahout.clustering.cdbw.TestCDbwEvaluator
  • org.apache.mahout.clustering.dirichlet.DirichletDriver
  • org.apache.mahout.clustering.dirichlet.DirichletMapper
  • org.apache.mahout.clustering.dirichlet.TestDirichletClustering
  • org.apache.mahout.clustering.dirichlet.TestL1ModelClustering
  • org.apache.mahout.clustering.dirichlet.TestMapReduce
  • org.apache.mahout.clustering.display.DisplayDirichlet
  • org.apache.mahout.clustering.syntheticcontrol.dirichlet.Job
  • org.apache.mahout.clustering.TestClusterDumper
  • org.apache.mahout.clustering.TestClusterEvaluator
    • 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.