Example usage for org.apache.mahout.clustering.classify ClusterClassifier ClusterClassifier

Introduction

In this page you can find the example usage for org.apache.mahout.clustering.classify ClusterClassifier ClusterClassifier.

Prototype

public ClusterClassifier(List<Cluster> models, ClusteringPolicy policy) 

Source Link

Document

The public constructor accepts a list of clusters to become the models

Usage

From source file:DisplayKMeans.java

License:Apache License

private static void runSequentialKMeansClassifier(Configuration conf, Path samples, Path output,
        DistanceMeasure measure, int numClusters, int maxIterations, double convergenceDelta)
        throws IOException {
    Collection<Vector> points = Lists.newArrayList();
    for (int i = 0; i < numClusters; i++) {
        points.add(SAMPLE_DATA.get(i).get());
        //      System.out.println(SAMPLE_DATA.get(i).toString());
    }// ww w . j  av  a  2  s  . co m
    List<Cluster> initialClusters = Lists.newArrayList();
    int id = 0;
    for (Vector point : points) {
        initialClusters.add(new org.apache.mahout.clustering.kmeans.Kluster(point, id++, measure));
    }
    ClusterClassifier prior = new ClusterClassifier(initialClusters,
            new KMeansClusteringPolicy(convergenceDelta));
    Path priorPath = new Path(output, Cluster.INITIAL_CLUSTERS_DIR);
    prior.writeToSeqFiles(priorPath);

    ClusterIterator.iterateSeq(conf, samples, priorPath, output, maxIterations);
    loadClustersWritable(output);
}

---

From source file:DisplayFuzzyKMeans.java

License:Apache License

private static void runSequentialFuzzyKClassifier(Configuration conf, Path samples, Path output,
        DistanceMeasure measure, int numClusters, int maxIterations, float m, double threshold)
        throws IOException {
    Collection<Vector> points = Lists.newArrayList();
    for (int i = 0; i < numClusters; i++) {
        points.add(SAMPLE_DATA.get(i).get());
    }/*w w  w.j a  va  2  s  .  co  m*/
    List<Cluster> initialClusters = Lists.newArrayList();
    int id = 0;
    for (Vector point : points) {
        initialClusters.add(new SoftCluster(point, id++, measure));
    }
    ClusterClassifier prior = new ClusterClassifier(initialClusters,
            new FuzzyKMeansClusteringPolicy(m, threshold));
    Path priorPath = new Path(output, "classifier-0");
    prior.writeToSeqFiles(priorPath);

    ClusterIterator.iterateSeq(conf, samples, priorPath, output, maxIterations);
    loadClustersWritable(output);
}

---

From source file:curation.mahout_test.DisplayKMeans.java

License:Apache License

private static void runSequentialKMeansClassifier(Configuration conf, Path samples, Path output,
        DistanceMeasure measure, int numClusters, int maxIterations, double convergenceDelta)
        throws IOException {
    Collection<Vector> points = Lists.newArrayList();
    for (int i = 0; i < numClusters; i++) {
        points.add(SAMPLE_DATA.get(i).get());
    }/*from  w ww  .  j a va  2s. com*/
    List<Cluster> initialClusters = Lists.newArrayList();
    int id = 0;
    for (Vector point : points) {
        initialClusters.add(new org.apache.mahout.clustering.kmeans.Kluster(point, id++, measure));
    }
    ClusterClassifier prior = new ClusterClassifier(initialClusters,
            new KMeansClusteringPolicy(convergenceDelta));
    Path priorPath = new Path(output, Cluster.INITIAL_CLUSTERS_DIR);
    prior.writeToSeqFiles(priorPath);

    ClusterIterator.iterateSeq(conf, samples, priorPath, output, maxIterations);
    loadClustersWritable(output);
}

---

From source file:org.aksw.tsoru.textmining.mahout.plot.Display.java

License:Apache License

private static void runSequentialFuzzyKClassifier(Configuration conf, Path samples, Path output,
        DistanceMeasure measure, int numClusters, int maxIterations, float m, double threshold)
        throws IOException {
    Collection<Vector> points = Lists.newArrayList();
    for (int i = 0; i < numClusters; i++) {
        points.add(SAMPLE_DATA.get(i).get());
    }/*  w w  w .j  ava2  s .c  om*/
    List<Cluster> initialClusters = Lists.newArrayList();
    int id = 0;
    for (Vector point : points) {
        initialClusters.add(new SoftCluster(point, id++, measure));
    }
    ClusterClassifier prior = new ClusterClassifier(initialClusters,
            new FuzzyKMeansClusteringPolicy(m, threshold));
    Path priorPath = new Path(output, "classifier-0");
    prior.writeToSeqFiles(priorPath);

    ClusterIterator.iterateSeq(conf, samples, priorPath, output, maxIterations);
    loadClustersWritable(output);

    prior.close();
}

---