Example usage for weka.core Instance value

List of usage examples for weka.core Instance value

  1. HOME
  2. Java
  3. weka
  4. weka.core.*
  5. Instance
  6. value

Introduction

In this page you can find the example usage for weka.core Instance value.

Prototype

public double value(Attribute att);

Source Link

Document

Returns an instance's attribute value in internal format.

Usage

From source file:moa.cluster.Riffle.java

License:Apache License

/**
 * Create a new cluster from an exemplar data point
 * @param x /*from  w ww  .j a  v  a  2 s .  com*/
 */
public Riffle(Instance x) {
    safeInit(x);
    this.numLabeledPoints = (int) Math.ceil(x.weight());
    this.labelFrequencies[(int) x.classValue()] += x.weight();
    this.gtLabelFrequencies[(int) x.classValue()]++;
    for (int i = 0; (i < this.symbolFrequencies.length) && (i < x.numAttributes()); ++i) {
        double value = x.value(i);
        if (this.symbolFrequencies[i] == null) {
            if ((this.parentClusterer != null) && (this.parentClusterer.getUniverse() != null)) {
                this.variances[i] = this.parentClusterer.getUniverse().variances[i];
            } else {
                this.variances[i] = this.initialStandardDeviationOption.getValue();
            }
        } else {
            this.variances[i] = 1;
            this.symbolFrequencies[i][(int) value]++;
        }
    }
    this.numTotalPoints = 1;
    this.setGroundTruth(x.classValue());
    this.setCenter(x.toDoubleArray());
    this.setWeight(x.weight());
    this.setRadius(this.initialStandardDeviationOption.getValue());
    this.runningSumOfSquares = 0.0;
    this.setId(autoindex.getAndIncrement());
}

From source file:moa.cluster.Riffle.java

License:Apache License

/**
 * Use the multi-variate Gaussian probability estimation approach to determine probability x is a member of this cluster.
 * This is not the same as a true Mahalanobis distance derived from a full covariance matrix as we take the heuristic shortcut
 * of assuming variable independence (rho = 0), and thus have an axis-parallel hyper-ellipse (i.e. diagonal covariance matrix)
 * under the same 'naive' assumption as in Naive Bayes. The computational speed-up here is worth the reduced accuracy.
 * @param x instance in question/*from   w  ww.j  a  v  a2s. co  m*/
 * @return probability that point x is a member of this cluster
 * @deprecated 
 */
@Deprecated
public double getInclusionMVGProbability(Instance x) {

    double[] diagonalCovariance = this.getVariances();
    int degreesOfFreedom = 0;
    double axisParallelSquaredMahalanobusDistance = 0;
    // Compute Mahalanobis (axis-parallel estimation) distance and tally degrees of freedom
    for (int i = 0; i < centroid.length; ++i) {
        if (i == x.classIndex()) {
            continue;
        }
        double attributeDelta = (this.symbolFrequencies[i] == null) ? x.value(i) - centroid[i]
                : (Math.abs(x.value(i) - centroid[i]) < 1e-32) ? 0 : 1;
        if ((diagonalCovariance[i] != 0) && Double.isFinite(diagonalCovariance[i])) {
            axisParallelSquaredMahalanobusDistance += attributeDelta * attributeDelta / diagonalCovariance[i];
        }
        degreesOfFreedom++;
    }

    // Zero distance means exact match, so stop working on more computations...
    if ((axisParallelSquaredMahalanobusDistance == 0) && (degreesOfFreedom > 0)) {
        return 1.0;
    }

    // Compute the determinant of the Covariance matrix (which is O(N) if we only have the diagonal of the matrix)
    double covarianceDet = 1;
    for (int i = 0; i < diagonalCovariance.length; ++i) {
        if (i == x.classIndex()) {
            continue;
        }
        if ((diagonalCovariance[i] != 0) && Double.isFinite(diagonalCovariance[i])) {
            covarianceDet *= diagonalCovariance[i];
        } else {
            covarianceDet *= weka.core.Utils.SMALL;
        }
    }
    if (covarianceDet == 0) {
        covarianceDet = weka.core.Utils.SMALL / 1000.0;
    } // Safety
    double score = Math.exp(0.0 - 0.5 * axisParallelSquaredMahalanobusDistance)
            / (Math.pow(2.0 * Math.PI, degreesOfFreedom / 2.0) * Math.sqrt(Math.abs(covarianceDet)));
    return (1.0 - weka.core.FastStats.chiSquaredProbability(score, degreesOfFreedom));
}

From source file:moa.cluster.SphereCluster.java

License:Apache License

public double getCenterDistance(Instance instance) {
    double distance = 0.0;
    //get the center through getCenter so subclass have a chance
    double[] center = getCenter();
    for (int i = 0; i < center.length; i++) {
        double d = center[i] - instance.value(i);
        distance += d * d;//from  www  .  j a  v  a 2s .  c  o m
    }
    return Math.sqrt(distance);
}

From source file:moa.clusterers.clustream.ClustreamKernel.java

License:Apache License

public void insert(Instance instance, long timestamp) {
    N++;/*from w  w  w .  j  a va 2s  . co m*/
    LST += timestamp;
    SST += timestamp * timestamp;

    for (int i = 0; i < instance.numValues(); i++) {
        LS[i] += instance.value(i);
        SS[i] += instance.value(i) * instance.value(i);
    }
}

From source file:moa.clusterers.clustream.ClustreamKernel.java

License:Apache License

/**
 * See interface <code>Cluster</code>
 * @param point/*  w ww.  ja v a 2 s  .com*/
 * @return
 */
@Override
public double getInclusionProbability(Instance instance) {
    //trivial cluster
    if (N == 1) {
        double distance = 0.0;
        for (int i = 0; i < LS.length; i++) {
            double d = LS[i] - instance.value(i);
            distance += d * d;
        }
        distance = Math.sqrt(distance);
        if (distance < EPSILON)
            return 1.0;
        return 0.0;
    } else {
        double dist = calcNormalizedDistance(instance.toDoubleArray());
        if (dist <= getRadius()) {
            return 1;
        } else {
            return 0;
        }
        //            double res = AuxiliaryFunctions.distanceProbabilty(dist, LS.length);
        //            return res;
    }
}

From source file:moa.clusterers.clustree.ClusKernel.java

License:Apache License

@Override
public double getInclusionProbability(Instance instance) {
    //trivial cluster
    if (N == 1) {
        double distance = 0.0;
        for (int i = 0; i < LS.length; i++) {
            double d = LS[i] - instance.value(i);
            distance += d * d;/*from   w  w  w .ja va  2s.co  m*/
        }
        distance = Math.sqrt(distance);
        if (distance < EPSILON)
            return 1.0;
        return 0.0;
    } else {
        double dist = calcNormalizedDistance(instance.toDoubleArray());
        if (dist <= getRadius()) {
            return 1;
        } else {
            return 0;
        }
        //            double res = AuxiliaryFunctions.distanceProbabilty(dist, LS.length);
        //            return res;
    }
}

From source file:moa.clusterers.denstream.MicroCluster.java

License:Apache License

public void insert(Instance instance, long timestamp) {
    N++;/* ww w .  ja v  a 2  s.  c  om*/
    super.setWeight(super.getWeight() + 1);
    this.lastEditT = timestamp;

    for (int i = 0; i < instance.numValues(); i++) {
        LS[i] += instance.value(i);
        SS[i] += instance.value(i) * instance.value(i);
    }
}

From source file:moa.clusterers.outliers.MyBaseOutlierDetector.java

License:Apache License

public double[] getInstanceValues(Instance inst) {
    double[] values = new double[inst.numValues() - 1]; // last attribute is the class
    for (int i = 0; i < inst.numValues() - 1; i++) {
        values[i] = inst.value(i);
    }//w  w  w .  j  av a 2  s.  c om
    return values;
}

From source file:moa.clusterers.outliers.MyBaseOutlierDetector.java

License:Apache License

public void PrintInstance(Instance inst) {
    Print("instance: [ ");
    for (int i = 0; i < inst.numValues() - 1; i++) { // last value is the class
        Printf("%.2f ", inst.value(i));
    }/*  ww w . ja v  a  2 s .c  o m*/
    Print("] ");
    Println("");
}

From source file:moa.core.utils.Converter.java

License:Open Source License

public List<Integer> getRelevantLabels(Instance x) {
    List<Integer> classValues = new LinkedList<Integer>();
    //get all class attributes
    for (int j = 0; j < m_L; j++) {
        if (x.value(j) > 0.0) {
            classValues.add(j);//from  ww  w.  ja  va2s  . com
        }
    }
    return classValues;
}