Examples of org.apache.mahout.clustering.Cluster

org.apache.mahout.clustering.Cluster
Implementations of this interface have a printable representation and certain attributes that are common across all clustering implementations

  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:{}",
          cluster.getId(), AbstractCluster.formatVector(cluster.getCenter(), null),
          cluster.getNumObservations(), AbstractCluster.formatVector(cluster.getRadius(), null));
      clusters.add(cluster);
    }
    return clusters;
  }

View Full Code Here

  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

    Path finalClustersPath = finalClustersPath(conf, clusterOutputPath);
    Iterator<?> it = new SequenceFileDirValueIterator<Writable>(finalClustersPath, PathType.LIST,
        PathFilters.partFilter(), null, false, conf);
    while (it.hasNext()) {
      ClusterWritable next = (ClusterWritable) it.next();
      Cluster cluster = next.getValue();
      cluster.configure(conf);
      clusterModels.add(cluster);
    }
    return clusterModels;
  }

View Full Code Here

    }
  }
  
  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

    }
  }
  
  private void write(VectorWritable vw, Context context, int clusterIndex, double weight)
    throws IOException, InterruptedException {
    Cluster cluster = clusterModels.get(clusterIndex);
    clusterId.set(cluster.getId());
    context.write(clusterId, new WeightedVectorWritable(weight, vw.get()));
  }

View Full Code Here

    Iterator<?> it = new SequenceFileDirValueIterator<Writable>(
        clusterFiles[0].getPath(), PathType.LIST, PathFilters.partFilter(),
        null, false, conf);
    while (it.hasNext()) {
      ClusterWritable next = (ClusterWritable) it.next();
      Cluster cluster = next.getValue();
      cluster.configure(conf);
      clusters.add(cluster);
    }
    return clusters;
  }

View Full Code Here

    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

    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

    this.subString = subString;
  }
  
  @Override
  public void write(ClusterWritable clusterWritable) throws IOException {
    Cluster cluster = clusterWritable.getValue();
    String fmtStr = cluster.asFormatString(dictionary);
    Writer writer = getWriter();
    if (subString > 0 && fmtStr.length() > subString) {
      writer.write(':');
      writer.write(fmtStr, 0, Math.min(subString, fmtStr.length()));
    } else {

View Full Code Here

  
  @Override
  protected void reduce(IntWritable key, Iterable<ClusterWritable> values, Context context) throws IOException,
      InterruptedException {
    Iterator<ClusterWritable> iter = values.iterator();
    Cluster first = iter.next().getValue(); // there must always be at least one
    while (iter.hasNext()) {
      Cluster cluster = iter.next().getValue();
      first.observe(cluster);
    }
    List<Cluster> models = Lists.newArrayList();
    models.add(first);
    classifier = new ClusterClassifier(models, policy);

View Full Code Here

0 1 2 3 4 5 6 7 8

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

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.