Examples of org.apache.mahout.clustering.canopy.Canopy

Package org.apache.mahout.clustering.canopy

Examples of org.apache.mahout.clustering.canopy.Canopy

org.apache.mahout.clustering.canopy.Canopy
This class models a canopy as a center point, the number of points that are contained within it according to the application of some distance metric, and a point total which is the sum of all the points and is used to compute the centroid when needed.

  }


  public void testCanopyAsFormatString() {
    double[] d = { 1.1, 2.2, 3.3 };
    Vector m = new DenseVector(d);
    Printable cluster = new Canopy(m, 123);
    String formatString = cluster.asFormatString(null);
    System.out.println(formatString);
    assertEquals("format", "C123: [1.100, 2.200, 3.300]", formatString);
  }

View Full Code Here


  public void testCanopyAsFormatStringSparse() {
    double[] d = { 1.1, 0.0, 3.3 };
    Vector m = new SequentialAccessSparseVector(3);
    m.assign(d);
    Printable cluster = new Canopy(m, 123);
    String formatString = cluster.asFormatString(null);
    System.out.println(formatString);
    assertEquals("format", "C123: [0:1.100, 2:3.300]", formatString);
  }

View Full Code Here

  }


  public void testCanopyAsFormatStringWithBindings() {
    double[] d = { 1.1, 2.2, 3.3 };
    Vector m = new DenseVector(d);
    Printable cluster = new Canopy(m, 123);
    String[] bindings = { "fee", null, null };
    String formatString = cluster.asFormatString(bindings);
    System.out.println(formatString);
    assertEquals("format", "C123: [fee:1.100, 1:2.200, 2:3.300]", formatString);
  }

View Full Code Here


  public void testCanopyAsFormatStringSparseWithBindings() {
    double[] d = { 1.1, 0.0, 3.3 };
    Vector m = new SequentialAccessSparseVector(3);
    m.assign(d);
    Printable cluster = new Canopy(m, 123);
    String formatString = cluster.asFormatString(null);
    System.out.println(formatString);
    assertEquals("format", "C123: [0:1.100, 2:3.300]", formatString);
  }

View Full Code Here


  @Test
  public void testCanopyAsFormatString() {
    double[] d = { 1.1, 2.2, 3.3 };
    Vector m = new DenseVector(d);
    Cluster cluster = new Canopy(m, 123, measure);
    String formatString = cluster.asFormatString(null);
    System.out.println(formatString);
    assertEquals("format", "C-123{n=0 c=[1.100, 2.200, 3.300] r=[]}", formatString);
  }

View Full Code Here

  @Test
  public void testCanopyAsFormatStringSparse() {
    double[] d = { 1.1, 0.0, 3.3 };
    Vector m = new SequentialAccessSparseVector(3);
    m.assign(d);
    Cluster cluster = new Canopy(m, 123, measure);
    String formatString = cluster.asFormatString(null);
    System.out.println(formatString);
    assertEquals("format", "C-123{n=0 c=[0:1.100, 2:3.300] r=[]}", formatString);
  }

View Full Code Here


  @Test
  public void testCanopyAsFormatStringWithBindings() {
    double[] d = { 1.1, 2.2, 3.3 };
    Vector m = new DenseVector(d);
    Cluster cluster = new Canopy(m, 123, measure);
    String[] bindings = { "fee", null, null };
    String formatString = cluster.asFormatString(bindings);
    System.out.println(formatString);
    assertEquals("format", "C-123{n=0 c=[fee:1.100, 1:2.200, 2:3.300] r=[]}", formatString);
  }

View Full Code Here

  @Test
  public void testCanopyAsFormatStringSparseWithBindings() {
    double[] d = { 1.1, 0.0, 3.3 };
    Vector m = new SequentialAccessSparseVector(3);
    m.assign(d);
    Cluster cluster = new Canopy(m, 123, measure);
    String formatString = cluster.asFormatString(null);
    System.out.println(formatString);
    assertEquals("format", "C-123{n=0 c=[0:1.100, 2:3.300] r=[]}", formatString);
  }

View Full Code Here


  @Test
  public void testCanopyClassification() {
    List<Model<VectorWritable>> models = new ArrayList<Model<VectorWritable>>();
    DistanceMeasure measure = new ManhattanDistanceMeasure();
    models.add(new Canopy(new DenseVector(2).assign(1), 0, measure));
    models.add(new Canopy(new DenseVector(2), 1, measure));
    models.add(new Canopy(new DenseVector(2).assign(-1), 2, measure));
    AbstractVectorClassifier classifier = new VectorModelClassifier(models);
    Vector pdf = classifier.classify(new DenseVector(2));
    assertEquals("[0,0]", "[0.107, 0.787, 0.107]", AbstractCluster.formatVector(pdf, null));
    pdf = classifier.classify(new DenseVector(2).assign(2));
    assertEquals("[2,2]", "[0.867, 0.117, 0.016]", AbstractCluster.formatVector(pdf, null));

View Full Code Here

   * @param dP a double representative point offset
   * @param measure the DistanceMeasure
   */
  private void initData(double dC, double dP, DistanceMeasure measure) {
    clusters = new ArrayList<Cluster>();
    clusters.add(new Canopy(new DenseVector(new double[] { -dC, -dC }), 1, measure));
    clusters.add(new Canopy(new DenseVector(new double[] { -dC, dC }), 3, measure));
    clusters.add(new Canopy(new DenseVector(new double[] { dC, dC }), 5, measure));
    clusters.add(new Canopy(new DenseVector(new double[] { dC, -dC }), 7, measure));
    representativePoints = new HashMap<Integer, List<VectorWritable>>();
    for (Cluster cluster : clusters) {
      List<VectorWritable> points = new ArrayList<VectorWritable>();
      representativePoints.put(cluster.getId(), points);
      points.add(new VectorWritable(cluster.getCenter().clone()));

View Full Code Here

0 1 2 3 4 5 6 7 8 9

TOP

Related Classes of org.apache.mahout.clustering.canopy.Canopy

org.apache.hadoop.io.Text

org.apache.mahout.clustering.cdbw.TestCDbwEvaluator

org.apache.mahout.clustering.fuzzykmeans.DisplayFuzzyKMeans

org.apache.mahout.clustering.fuzzykmeans.FuzzyKMeansUtil

org.apache.mahout.clustering.iterator.TestClusterClassifier

org.apache.mahout.clustering.kmeans.DisplayKMeans

org.apache.mahout.clustering.kmeans.KMeansUtil

org.apache.mahout.clustering.syntheticcontrol.canopy.OutputMapper

org.apache.mahout.clustering.TestClusterClassifier

org.apache.mahout.clustering.TestClusterEvaluator

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.