Package org.apache.mahout.clustering

Examples of org.apache.mahout.clustering.Cluster

233
234
235
236
237
238
239
240
    }
  }
 
  private static void write(List<Cluster> clusterModels, SequenceFile.Writer writer, WeightedVectorWritable wvw,
      int maxValueIndex) throws IOException {
    Cluster cluster = clusterModels.get(maxValueIndex);
    writer.append(new IntWritable(cluster.getId()), wvw);
  }
View Full Code Here
111
112
113
114
115
116
117
118
119
120
    }
  }
 
  private void write(VectorWritable vw, Context context, int clusterIndex)
      throws IOException, InterruptedException {
    Cluster cluster = clusterModels.get(clusterIndex);
    clusterId.set(cluster.getId());
    weightedVW.setVector(vw.get());
    context.write(clusterId, weightedVW);
  }
View Full Code Here
120
121
122
123
124
125
126
127
128
129
130
131
132
133
134
135
136
137
138
139
  }
 
  public static List<Cluster> populateClusterModels(Path clusterOutputPath, Configuration conf)
      throws IOException {
    List<Cluster> clusters = new ArrayList<Cluster>();
    Cluster cluster = null;
    FileSystem fileSystem = clusterOutputPath.getFileSystem(conf);
    FileStatus[] clusterFiles = fileSystem.listStatus(clusterOutputPath,
        PathFilters.finalPartFilter());
    Iterator<?> it = new SequenceFileDirValueIterator<Writable>(
        clusterFiles[0].getPath(), PathType.LIST, PathFilters.partFilter(),
        null, false, conf);
    while (it.hasNext()) {
      ClusterWritable next = (ClusterWritable) it.next();
      cluster = next.getValue();
      cluster.configure(conf);
      clusters.add(cluster);
    }
    return clusters;
  }
View Full Code Here
131
132
133
134
135
136
137
138
139
140
    models = Lists.newArrayList();
    ClusteringPolicyWritable clusteringPolicyWritable = new ClusteringPolicyWritable();
    clusteringPolicyWritable.readFields(in);
    policy = clusteringPolicyWritable.getValue();
    for (int i = 0; i < size; i++) {
      Cluster element = ClassUtils.instantiateAs(modelClass, Cluster.class);
      element.readFields(in);
      models.add(element);
    }
  }
View Full Code Here
187
188
189
190
191
192
193
194
195
196
197
    FileSystem fs = FileSystem.get(path.toUri(), config);
    SequenceFile.Writer writer = null;
    ClusterWritable cw = new ClusterWritable();
    for (int i = 0; i < models.size(); i++) {
      try {
        Cluster cluster = models.get(i);
        cw.setValue(cluster);
        writer = new SequenceFile.Writer(fs, config,
            new Path(path, "part-" + String.format(Locale.ENGLISH, "%05d", i)), IntWritable.class,
            ClusterWritable.class);
        Writable key = new IntWritable(i);
View Full Code Here
205
206
207
208
209
210
211
212
213
214
215
216
  public void readFromSeqFiles(Configuration conf, Path path) throws IOException {
    Configuration config = new Configuration();
    List<Cluster> clusters = Lists.newArrayList();
    for (ClusterWritable cw : new SequenceFileDirValueIterable<ClusterWritable>(path, PathType.LIST,
        PathFilters.logsCRCFilter(), config)) {
      Cluster cluster = cw.getValue();
      cluster.configure(conf);
      clusters.add(cluster);
    }
    this.models = clusters;
    modelClass = models.get(0).getClass().getName();
    this.policy = readPolicy(path);
View Full Code Here
331
332
333
334
335
336
337
338
339
340
341
342
343
344
  protected static List<Cluster> readClustersWritable(Path clustersIn) {
    List<Cluster> clusters = Lists.newArrayList();
    Configuration conf = new Configuration();
    for (ClusterWritable value : new SequenceFileDirValueIterable<ClusterWritable>(clustersIn, PathType.LIST,
        PathFilters.logsCRCFilter(), conf)) {
      Cluster cluster = value.getValue();
      log.info(
          "Reading Cluster:{} center:{} numPoints:{} radius:{}",
          new Object[] {cluster.getId(), AbstractCluster.formatVector(cluster.getCenter(), null),
              cluster.getNumObservations(), AbstractCluster.formatVector(cluster.getRadius(), null)});
      clusters.add(cluster);
    }
    return clusters;
  }
View Full Code Here
295
296
297
298
299
300
301
302
303
304
305
306
    int row = 0;
    StringBuilder models = new StringBuilder(100);
    for (List<Cluster> r : result) {
      models.append("sample[").append(row++).append("]= ");
      for (int k = 0; k < r.size(); k++) {
        Cluster model = r.get(k);
        models.append('m').append(k).append(model.asFormatString(null)).append(", ");
      }
      models.append('\n');
    }
    models.append('\n');
    System.out.println(models.toString());
View Full Code Here
94
95
96
97
98
99
100
101
102
103
104
105
106
107
108
109
110
111
112
113
114
115
116
117
118
119
120
121
122
123
124
125
126
127
128
   */

  @Override
  public void write(ClusterWritable clusterWritable) throws IOException {
    StringBuilder line = new StringBuilder();
    Cluster cluster = clusterWritable.getValue();
  Color rgb = getColor(cluster.getId());

    String topTerms = "";
    if (dictionary != null) {
      topTerms = getTopTerms(cluster.getCenter(), dictionary, numTopFeatures);
    }
    String clusterLabel = String.valueOf(cluster.getId()) + "_" + topTerms;
    //do some positioning so that items are visible and grouped together
    //TODO: put in a real layout algorithm
    float x = lastX + 1000;
    float y = lastY;
    if (x > (1000 + posStep)) {
      y = lastY + 1000;
      x = 0;
    }

    line.append(createNode(clusterLabel, rgb, x, y));
    List<WeightedVectorWritable> points = clusterIdToPoints.get(cluster.getId());
    if (points != null) {
      for (WeightedVectorWritable point : points) {
        Vector theVec = point.getVector();
        double distance = 1;
        if (measure != null) {
          distance = measure.distance(cluster.getCenter().getLengthSquared(), cluster.getCenter(), theVec) * 500; //scale the distance
        }
        String vecStr;
        int angle = random.nextInt(360); //pick an angle at random and then scale along that angle
        double angleRads = Math.toRadians(angle);

View Full Code Here
44
45
46
47
48
49
50
51
52
53
54
55
56
  }

  @Override
  public void write(ClusterWritable clusterWritable) throws IOException {
    StringBuilder line = new StringBuilder();
    Cluster cluster = clusterWritable.getValue();
  line.append(cluster.getId());
    List<WeightedVectorWritable> points = getClusterIdToPoints().get(cluster.getId());
    if (points != null) {
      for (WeightedVectorWritable point : points) {
        Vector theVec = point.getVector();
        line.append(',');
        if (theVec instanceof NamedVector) {
View Full Code Here
TOP

Related Classes of org.apache.mahout.clustering.Cluster

  • com.ikanow.infinit.e.processing.custom.utils.HadoopUtils
  • mia.clustering.ch10.InterClusterDistances
  • org.apache.mahout.clustering.cdbw.CDbwEvaluator
  • org.apache.mahout.clustering.classify.ClusterClassificationDriver
  • org.apache.mahout.clustering.classify.ClusterClassificationMapper
  • org.apache.mahout.clustering.classify.ClusterClassifier
  • org.apache.mahout.clustering.dirichlet.DirichletCluster
  • org.apache.mahout.clustering.dirichlet.DirichletReducer
  • org.apache.mahout.clustering.dirichlet.JsonClusterAdapter
  • org.apache.mahout.clustering.dirichlet.TestL1ModelClustering
    • 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.