Examples of EuclideanDistanceMeasure

org.apache.mahout.common.distance.EuclideanDistanceMeasure
This class implements a Euclidean distance metric by summing the square root of the squared differences between each coordinate.
If you don't care about the true distance and only need the values for comparison, then the base class, {@link SquaredEuclideanDistanceMeasure}, will be faster since it doesn't do the actual square root of the squared differences.
org.apache.mahout.utils.EuclideanDistanceMeasure
This class implements a Euclidian distance metric by summing the square root of the squared differences between each coordinate

Examples of org.apache.mahout.common.distance.EuclideanDistanceMeasure

    }
    System.out.printf("Generated query matrix.\n");


    for (int threads : new int[]{1, 2, 3, 4, 5, 6, 10, 20, 50}) {
      for (int block : new int[]{1, 10, 50}) {
        BruteSearch search = new BruteSearch(new EuclideanDistanceMeasure());
        search.addAll(referenceVectors);
        long t0 = System.nanoTime();
        search.search(queryVectors, block, threads);
        long t1 = System.nanoTime();
        System.out.printf("%d\t%d\t%.2f\n", threads, block, (t1 - t0) / 1e9);

View Full Code Here

Examples of org.apache.mahout.common.distance.EuclideanDistanceMeasure

    SequenceFile.Writer writer = new SequenceFile.Writer(fs, conf,
        path, Text.class, Cluster.class);
    
    for (int i = 0; i < k; i++) {
      Vector vec = vectors.get(i);
      Cluster cluster = new Cluster(vec, i, new EuclideanDistanceMeasure());
      writer.append(new Text(cluster.getIdentifier()), cluster);
    }
    writer.close();
    
    KMeansDriver.run(conf, new Path("testdata/points"), new Path("testdata/clusters"),
      new Path("output"), new EuclideanDistanceMeasure(), 0.001, 10,
      true, false);
    
    SequenceFile.Reader reader = new SequenceFile.Reader(fs,
        new Path("output/" + Cluster.CLUSTERED_POINTS_DIR
                 + "/part-m-00000"), conf);

View Full Code Here

Examples of org.apache.mahout.common.distance.EuclideanDistanceMeasure

    List<Vector> randomPoints = RandomPointsUtil.chooseRandomPoints(sampleData,  k);
    List<SoftCluster> clusters = new ArrayList<SoftCluster>();


    int clusterId = 0;
    for (Vector v : randomPoints) {
      clusters.add(new SoftCluster(v, clusterId++, new EuclideanDistanceMeasure()));
    }


    List<List<SoftCluster>> finalClusters = FuzzyKMeansClusterer
        .clusterPoints(sampleData, clusters,
            new EuclideanDistanceMeasure(), 0.01, 3, 10);
    for (SoftCluster cluster : finalClusters.get(finalClusters.size() - 1)) {
      System.out.println("Fuzzy Cluster id: " + cluster.getId()
          + " center: " + cluster.getCenter().asFormatString());
    }
  }

View Full Code Here

Examples of org.apache.mahout.common.distance.EuclideanDistanceMeasure

        sampleData, k);
    List<Cluster> clusters = new ArrayList<Cluster>();


    int clusterId = 0;
    for (Vector v : randomPoints) {
      clusters.add(new Cluster(v, clusterId++, new EuclideanDistanceMeasure()));
    }


    List<List<Cluster>> finalClusters = KMeansClusterer.clusterPoints(
        sampleData, clusters, new EuclideanDistanceMeasure(), 3, 0.01);
    for (Cluster cluster : finalClusters.get(finalClusters.size() - 1)) {
      System.out.println("Cluster id: " + cluster.getId() + " center: "
          + cluster.getCenter().asFormatString());
    }
  }

View Full Code Here

Examples of org.apache.mahout.common.distance.EuclideanDistanceMeasure

    Path vectorsFolder = new Path(outputDir, "tfidf-vectors");
    Path canopyCentroids = new Path(outputDir , "canopy-centroids");
    Path clusterOutput = new Path(outputDir , "clusters");
    
    CanopyDriver.run(vectorsFolder, canopyCentroids,
      new EuclideanDistanceMeasure(), 250, 120, false, false);
    KMeansDriver.run(conf, vectorsFolder, new Path(canopyCentroids, "clusters-0"),
      clusterOutput, new TanimotoDistanceMeasure(), 0.01,
      20, true, false);
    
    SequenceFile.Reader reader = new SequenceFile.Reader(fs,

View Full Code Here

Examples of org.apache.mahout.common.distance.EuclideanDistanceMeasure

    RandomPointsUtil.generateSamples(sampleData, 400, 1, 1, 2);
    RandomPointsUtil.generateSamples(sampleData, 300, 1, 0, 0.5);
    RandomPointsUtil.generateSamples(sampleData, 300, 0, 2, 0.1);


    List<Canopy> canopies = CanopyClusterer.createCanopies(sampleData,
        new EuclideanDistanceMeasure(), 3.0, 1.5);


    for (Canopy canopy : canopies) {
      System.out.println("Canopy id: " + canopy.getId() + " center: "
          + canopy.getCenter().asFormatString());
    }

View Full Code Here

Examples of org.apache.mahout.common.distance.EuclideanDistanceMeasure

  /**
   * Story: User can exercise the reference implementation to verify that the test datapoints are clustered in
   * a reasonable manner.
   */
  public void testReferenceImplementation() {
    MeanShiftCanopyClusterer clusterer = new MeanShiftCanopyClusterer(new EuclideanDistanceMeasure(), 4.0,
        1.0, 0.5);
    List<MeanShiftCanopy> canopies = new ArrayList<MeanShiftCanopy>();
    // add all points to the canopies
    int nextCanopyId = 0;
    for (Vector aRaw : raw) {

View Full Code Here

Examples of org.apache.mahout.common.distance.EuclideanDistanceMeasure

        clusterList.add(cluster);
      }
      Map<String,String> pointClusterInfo = new HashMap<String,String>();
      // run reference FuzzyKmeans algorithm
      List<List<SoftCluster>> clusters = FuzzyKMeansClusterer.clusterPoints(points, clusterList,
        new EuclideanDistanceMeasure(), 0.001, 2, 2);
      computeCluster(points, clusters.get(clusters.size() - 1), new FuzzyKMeansClusterer(
          new EuclideanDistanceMeasure(), 0.001, 2), pointClusterInfo);
      
      // iterate for each point
      for (String value : pointClusterInfo.values()) {
        String clusterInfoStr = value.substring(1, value.length() - 1);
        String[] clusterInfoList = clusterInfoStr.split(" ");

View Full Code Here

Examples of org.apache.mahout.common.distance.EuclideanDistanceMeasure

  }
  
  /** Story: Test the reference implementation */
  public void testReferenceImplementation() throws Exception {
    List<Vector> points = getPoints(reference);
    DistanceMeasure measure = new EuclideanDistanceMeasure();
    // try all possible values of k
    for (int k = 0; k < points.size(); k++) {
      System.out.println("Test k=" + (k + 1) + ':');
      // pick k initial cluster centers at random
      List<Cluster> clusters = new ArrayList<Cluster>();

View Full Code Here

Examples of org.apache.mahout.common.distance.EuclideanDistanceMeasure

      for (VectorWritable point : points) {
        mapper.map(new Text(), point, collector, null);
      }
      assertEquals("Number of map results", k + 1, collector.getData().size());
      // now verify that all points are correctly allocated
      EuclideanDistanceMeasure euclideanDistanceMeasure = new EuclideanDistanceMeasure();
      Map<String,Cluster> clusterMap = loadClusterMap(clusters);
      for (String key : collector.getKeys()) {
        Cluster cluster = clusterMap.get(key);
        List<KMeansInfo> values = collector.getValue(key);
        for (KMeansInfo value : values) {
          double distance = euclideanDistanceMeasure.distance(cluster.getCenter(), value.getPointTotal());
          for (Cluster c : clusters) {
            assertTrue("distance error", distance <= euclideanDistanceMeasure.distance(value.getPointTotal(),
              c.getCenter()));
          }
        }
      }
    }

View Full Code Here

0 1 2 3 4 5

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.