Examples of SampledNormalDistribution

org.apache.mahout.clustering.dirichlet.models.SampledNormalDistribution
An implementation of the ModelDistribution interface suitable for testing the DirichletCluster algorithm. Uses a Normal Distribution to sample the prior model values.

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

  public void testReducer() throws Exception {
    generateSamples(100, 0, 0, 1);
    generateSamples(100, 2, 0, 1);
    generateSamples(100, 0, 2, 1);
    generateSamples(100, 2, 2, 1);
    DirichletState state = new DirichletState(new SampledNormalDistribution(new VectorWritable(new DenseVector(2))), 20, 1);
    DirichletMapper mapper = new DirichletMapper();
    mapper.setup(state);


    DummyRecordWriter<Text, VectorWritable> mapWriter = new DummyRecordWriter<Text, VectorWritable>();
    Mapper<WritableComparable<?>, VectorWritable, Text, VectorWritable>.Context mapContext = DummyRecordWriter.build(mapper,

View Full Code Here

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

  public void testMRIterations() throws Exception {
    generateSamples(100, 0, 0, 1);
    generateSamples(100, 2, 0, 1);
    generateSamples(100, 0, 2, 1);
    generateSamples(100, 2, 2, 1);
    DirichletState state = new DirichletState(new SampledNormalDistribution(new VectorWritable(new DenseVector(2))), 20, 1.0);


    Collection<Model<VectorWritable>[]> models = new ArrayList<Model<VectorWritable>[]>();


    for (int iteration = 0; iteration < 10; iteration++) {
      DirichletMapper mapper = new DirichletMapper();

View Full Code Here

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

    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;
    AbstractVectorModelDistribution modelDistribution = new SampledNormalDistribution(new VectorWritable(new DenseVector(2)));
    String[] args = { optKey(DefaultOptionCreator.INPUT_OPTION), getTestTempDirPath("input").toString(),
        optKey(DefaultOptionCreator.OUTPUT_OPTION), getTestTempDirPath("output").toString(),
        optKey(DirichletDriver.MODEL_DISTRIBUTION_CLASS_OPTION), modelDistribution.getClass().getName(),
        optKey(DirichletDriver.MODEL_PROTOTYPE_CLASS_OPTION), modelDistribution.getModelPrototype().get().getClass().getName(),
        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 };
    new DirichletDriver().run(args);
    // and inspect results
    Collection<List<DirichletCluster>> clusters = new ArrayList<List<DirichletCluster>>();
    Configuration conf = new Configuration();
    conf.set(DirichletDriver.MODEL_DISTRIBUTION_KEY, modelDistribution.asJsonString());
    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

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

    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;
    AbstractVectorModelDistribution modelDistribution = new SampledNormalDistribution(new VectorWritable(new DenseVector(2)));
    String[] args = { optKey(DefaultOptionCreator.INPUT_OPTION), getTestTempDirPath("input").toString(),
        optKey(DefaultOptionCreator.OUTPUT_OPTION), getTestTempDirPath("output").toString(),
        optKey(DirichletDriver.MODEL_DISTRIBUTION_CLASS_OPTION), modelDistribution.getClass().getName(),
        optKey(DirichletDriver.MODEL_PROTOTYPE_CLASS_OPTION), modelDistribution.getModelPrototype().get().getClass().getName(),
        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 = new ArrayList<List<DirichletCluster>>();
    Configuration conf = new Configuration();
    conf.set(DirichletDriver.MODEL_DISTRIBUTION_KEY, modelDistribution.asJsonString());
    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

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

  @Test
  public void testDriverMnRIterations() throws Exception {
    generate4Datasets();
    // Now run the driver
    int maxIterations = 3;
    AbstractVectorModelDistribution modelDistribution = new SampledNormalDistribution(new VectorWritable(new DenseVector(2)));
    Configuration conf = new Configuration();
    DirichletDriver.run(conf,
                        getTestTempDirPath("input"),
                        getTestTempDirPath("output"),
                        modelDistribution,
                        20,
                        maxIterations,
                        1.0,
                        false,
                        true,
                        0,
                        false);
    // and inspect results
    List<List<DirichletCluster>> clusters = new ArrayList<List<DirichletCluster>>();
    conf.set(DirichletDriver.MODEL_DISTRIBUTION_KEY, modelDistribution.asJsonString());
    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

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

    generateResults();
    new DisplaySNDirichlet();
  }


  static void generateResults() {
    DisplayDirichlet.generateResults(new SampledNormalDistribution());
  }

View Full Code Here

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

    generateSamples(40, 1, 1, 3);
    generateSamples(30, 1, 0, 0.1);
    generateSamples(30, 0, 1, 0.1);


    DirichletClusterer<Vector> dc = new DirichletClusterer<Vector>(sampleData,
        new SampledNormalDistribution(), 1.0, 10, 1, 0);
    List<Model<Vector>[]> result = dc.cluster(30);
    printResults(result, 2);
    assertNotNull(result);
  }

View Full Code Here

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

    generateSamples(400, 1, 1, 3);
    generateSamples(300, 1, 0, 0.1);
    generateSamples(300, 0, 1, 0.1);


    DirichletClusterer<Vector> dc = new DirichletClusterer<Vector>(sampleData,
        new SampledNormalDistribution(), 1.0, 10, 1, 0);
    List<Model<Vector>[]> result = dc.cluster(30);
    printResults(result, 20);
    assertNotNull(result);
  }

View Full Code Here

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

    generateSamples(4000, 1, 1, 3);
    generateSamples(3000, 1, 0, 0.1);
    generateSamples(3000, 0, 1, 0.1);


    DirichletClusterer<Vector> dc = new DirichletClusterer<Vector>(sampleData,
        new SampledNormalDistribution(), 1.0, 10, 1, 0);
    List<Model<Vector>[]> result = dc.cluster(30);
    printResults(result, 200);
    assertNotNull(result);
  }

View Full Code Here

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

    generateSamples(100, 0, 0, 1);
    generateSamples(100, 2, 0, 1);
    generateSamples(100, 0, 2, 1);
    generateSamples(100, 2, 2, 1);
    DirichletState<Vector> state = new DirichletState<Vector>(
        new SampledNormalDistribution(), 20, 1, 1, 0);
    DirichletMapper mapper = new DirichletMapper();
    mapper.configure(state);


    DummyOutputCollector<Text, Text> mapCollector = new DummyOutputCollector<Text, Text>();
    for (Vector v : sampleData)

View Full Code Here

0 1 2 3

TOP

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.