Example usage for org.apache.mahout.math Vector getLengthSquared

Introduction

In this page you can find the example usage for org.apache.mahout.math Vector getLengthSquared.

Prototype

double getLengthSquared();

Source Link

Document

Return the sum of squares of all elements in the vector.

Usage

From source file:com.cloudera.science.ml.kmeans.parallel.CentersIndex.java

License:Open Source License

public void add(Vector vec, int centerId) {
    points.get(centerId).add(Vectors.toArray(vec));
    lengthSquared.get(centerId).add(vec.getLengthSquared());
    pointsPerCenter[centerId]++;/*from  w w w .j  a  v  a  2  s  .c o m*/
    updated = true;
}

From source file:com.cloudera.science.ml.kmeans.parallel.CentersIndex.java

License:Open Source License

public Distances getDistances(Vector vec, boolean approx) {
    int[] closestPoints = new int[pointsPerCenter.length];
    double[] distances = new double[pointsPerCenter.length];

    if (approx) {
        if (updated) {
            buildIndices();/* w  w  w  .j  a  v  a  2  s .  c om*/
        }

        BitSet q = index(vec);
        for (int i = 0; i < pointsPerCenter.length; i++) {
            List<BitSet> index = indices.get(i);
            SortedSet<Idx> lookup = Sets.newTreeSet();
            for (int j = 0; j < index.size(); j++) {
                Idx idx = new Idx(hammingDistance(q, index.get(j)), j);
                if (lookup.size() < projectionSamples) {
                    lookup.add(idx);
                } else if (idx.compareTo(lookup.last()) < 0) {
                    lookup.add(idx);
                    lookup.remove(lookup.last());
                }
            }

            List<double[]> p = points.get(i);
            distances[i] = Double.POSITIVE_INFINITY;
            for (Idx idx : lookup) {
                double lenSq = lengthSquared.get(i).get(idx.index);
                double d = vec.getLengthSquared() + lenSq - 2 * dot(vec, p.get(idx.index));
                if (d < distances[i]) {
                    distances[i] = d;
                    closestPoints[i] = idx.index;
                }
            }
        }
    } else { // More expensive exact computation
        for (int i = 0; i < pointsPerCenter.length; i++) {
            distances[i] = Double.POSITIVE_INFINITY;
            List<double[]> px = points.get(i);
            List<Double> lsq = lengthSquared.get(i);
            for (int j = 0; j < px.size(); j++) {
                double[] p = px.get(j);
                double lenSq = lsq.get(j);
                double d = vec.getLengthSquared() + lenSq - 2 * dot(vec, p);
                if (d < distances[i]) {
                    distances[i] = d;
                    closestPoints[i] = j;
                }
            }
        }
    }

    return new Distances(distances, closestPoints);
}

From source file:com.scaleunlimited.classify.vectors.UnitNormalizer.java

License:Apache License

@Override
public void normalize(Vector vector) {

    double length = Math.sqrt(vector.getLengthSquared());

    // Divide each vector coordinate by length, so we wind up with a unit vector.
    int cardinality = vector.size();
    for (int j = 0; j < cardinality; j++) {
        double curValue = vector.getQuick(j);
        if (curValue > 0.0) {
            vector.setQuick(j, curValue / length);
        }//from  w w  w . j  av a 2 s.c o  m
    }
}

From source file:edu.indiana.d2i.htrc.kmeans.KMeansClusterer.java

License:Apache License

/**
 * Iterates over all clusters and identifies the one closes to the given point. Distance measure used is
 * configured at creation time.//from w  w  w  .  jav  a  2 s  .  co m
 * 
 * @param point
 *          a point to find a cluster for.
 * @param clusters
 *          a List<Cluster> to test.
 */
public void emitPointToNearestCluster(Vector point, Iterable<Cluster> clusters,
        Mapper<?, ?, Text, ClusterObservations>.Context context) throws IOException, InterruptedException {
    Cluster nearestCluster = null;
    double nearestDistance = Double.MAX_VALUE;
    for (Cluster cluster : clusters) {
        Vector clusterCenter = cluster.getCenter();
        double distance = this.measure.distance(clusterCenter.getLengthSquared(), clusterCenter, point);
        if (log.isDebugEnabled()) {
            log.debug("{} Cluster: {}", distance, cluster.getId());
        }
        if (distance < nearestDistance || nearestCluster == null) {
            nearestCluster = cluster;
            nearestDistance = distance;
        }
    }
    context.write(new Text(nearestCluster.getIdentifier()),
            new ClusterObservations(1, point, point.times(point)));
}

From source file:edu.indiana.d2i.htrc.kmeans.KMeansClusterer.java

License:Apache License

public void outputPointWithClusterInfo(Vector vector, Iterable<Cluster> clusters,
        Mapper<?, ?, IntWritable, WeightedPropertyVectorWritable>.Context context)
        throws IOException, InterruptedException {
    AbstractCluster nearestCluster = null;
    double nearestDistance = Double.MAX_VALUE;
    for (AbstractCluster cluster : clusters) {
        Vector clusterCenter = cluster.getCenter();
        double distance = measure.distance(clusterCenter.getLengthSquared(), clusterCenter, vector);
        if (distance < nearestDistance || nearestCluster == null) {
            nearestCluster = cluster;//  w w w  .j a v a 2  s .c o m
            nearestDistance = distance;
        }
    }
    Map<Text, Text> props = new HashMap<Text, Text>();
    props.put(new Text("distance"), new Text(String.valueOf(nearestDistance)));
    context.write(new IntWritable(nearestCluster.getId()),
            new WeightedPropertyVectorWritable(1, vector, props));
}

From source file:edu.indiana.d2i.htrc.kmeans.KMeansClusterer.java

License:Apache License

/**
 * Iterates over all clusters and identifies the one closes to the given point. Distance measure used is
 * configured at creation time.//from   w ww .  ja va  2s  .  c o  m
 * 
 * @param point
 *          a point to find a cluster for.
 * @param clusters
 *          a List<Cluster> to test.
 */
protected void emitPointToNearestCluster(Vector point, Iterable<Cluster> clusters, Writer writer)
        throws IOException {
    AbstractCluster nearestCluster = null;
    double nearestDistance = Double.MAX_VALUE;
    for (AbstractCluster cluster : clusters) {
        Vector clusterCenter = cluster.getCenter();
        double distance = this.measure.distance(clusterCenter.getLengthSquared(), clusterCenter, point);
        if (log.isDebugEnabled()) {
            log.debug("{} Cluster: {}", distance, cluster.getId());
        }
        if (distance < nearestDistance || nearestCluster == null) {
            nearestCluster = cluster;
            nearestDistance = distance;
        }
    }
    writer.append(new IntWritable(nearestCluster.getId()), new WeightedVectorWritable(1, point));
}