Examples of MeanVariance

InfoCollection.util.MeanVariance
de.lmu.ifi.dbs.elki.math.MeanVariance
Do some simple statistics (mean, variance) using a numerically stable online algorithm. This class can repeatedly be fed with data using the add() methods, the resulting values for mean and average can be queried at any time using getMean() and getSampleVariance(). Make sure you have understood variance correctly when using getNaiveVariance() - since this class is fed with samples and estimates the mean from the samples, getSampleVariance() is the proper formula. Trivial code, but replicated a lot. The class is final so it should come at low cost. Related Literature:
B. P. Welford
Note on a method for calculating corrected sums of squares and products
in: Technometrics 4(3)

D.H.D. West
Updating Mean and Variance Estimates: An Improved Method
In: Communications of the ACM, Volume 22 Issue 9
@author Erich Schubert

Examples of InfoCollection.util.MeanVariance

   * repeats the experiment, returning a sample average. The other runs the
   * experiment once and returns the loss realized.
   */
  public static double LossAtEnd(SamplingRule policy, Truth truth, int N,
      int nruns) throws Exception {
    MeanVariance lossEstimator = new MeanVariance();
    Random rnd = new Random();
    for (int r = 0; r < nruns; r++) {
      policy.Start(truth.M());
      for (int n = 0; n < N; n++) {
        int x = policy.GetMeasurementDecision();
        double y = truth.Sample(x, rnd);
        policy.RecordMeasurement(x, y);
      }
      int i = policy.GetImplementationDecision();
      lossEstimator.AddSample(truth.Loss(i));
    }
    return lossEstimator.SampleMean();
  }

View Full Code Here

Examples of de.lmu.ifi.dbs.elki.math.MeanVariance

  }


  @Override
  public void prepare(OutlierResult or) {
    if(fixedmean == null) {
      MeanVariance mv = new MeanVariance();
      for(DBID id : or.getScores().iterDBIDs()) {
        double val = or.getScores().get(id);
        if(!Double.isNaN(val) && !Double.isInfinite(val)) {
          mv.put(val);
        }
      }
      mean = mv.getMean();
      factor = lambda * mv.getSampleStddev() * MathUtil.SQRT2;
      if (factor == 0.0) {
        factor = Double.MIN_NORMAL;
      }
    }
    else {

View Full Code Here

Examples of de.lmu.ifi.dbs.elki.math.MeanVariance

        }
      }
      min = mm.getMin();
    }
    if(mean == null) {
      MeanVariance mv = new MeanVariance();
      for(DBID id : or.getScores().iterDBIDs()) {
        double val = or.getScores().get(id);
        val = (val <= min) ? 0 : Math.sqrt(val - min);
        mv.put(val);
      }
      mean = mv.getMean();
      factor = lambda * mv.getSampleStddev() * MathUtil.SQRT2;
    }
    else {
      double sqsum = 0;
      int cnt = 0;
      for(DBID id : or.getScores().iterDBIDs()) {

View Full Code Here

Examples of de.lmu.ifi.dbs.elki.math.MeanVariance

    // Euclidean
    {
      DistanceQuery<NumberVector<?, ?>, DoubleDistance> dq = db.getDistanceQuery(relation, EuclideanDistanceFunction.STATIC);
      KNNQuery<NumberVector<?, ?>, DoubleDistance> knnq = QueryUtil.getLinearScanKNNQuery(dq);


      MeanVariance meansize = new MeanVariance();
      for(DBID id : relation.iterDBIDs()) {
        List<DistanceResultPair<DoubleDistance>> knnlist = knnq.getKNNForDBID(id, 2);
        meansize.put(knnlist.size());
      }
      org.junit.Assert.assertEquals("Euclidean mean 2NN", mean2nnEuclid, meansize.getMean(), 0.00001);
      org.junit.Assert.assertEquals("Euclidean variance 2NN", var2nnEuclid, meansize.getSampleVariance(), 0.00001);
    }
    // Manhattan
    {
      DistanceQuery<NumberVector<?, ?>, DoubleDistance> dq = db.getDistanceQuery(relation, ManhattanDistanceFunction.STATIC);
      KNNQuery<NumberVector<?, ?>, DoubleDistance> knnq = QueryUtil.getLinearScanKNNQuery(dq);


      MeanVariance meansize = new MeanVariance();
      for(DBID id : relation.iterDBIDs()) {
        List<DistanceResultPair<DoubleDistance>> knnlist = knnq.getKNNForDBID(id, 2);
        meansize.put(knnlist.size());
      }
      org.junit.Assert.assertEquals("Manhattan mean 2NN", mean2nnManhattan, meansize.getMean(), 0.00001);
      org.junit.Assert.assertEquals("Manhattan variance 2NN", var2nnManhattan, meansize.getSampleVariance(), 0.00001);
    }
  }

View Full Code Here

Examples of de.lmu.ifi.dbs.elki.math.MeanVariance

    // Euclidean
    {
      KNNJoin<DoubleVector, DoubleDistance, ?, ?> knnjoin = new KNNJoin<DoubleVector, DoubleDistance, RStarTreeNode, SpatialEntry>(EuclideanDistanceFunction.STATIC, 2);
      DataStore<KNNList<DoubleDistance>> result = knnjoin.run(db);


      MeanVariance meansize = new MeanVariance();
      for(DBID id : relation.getDBIDs()) {
        KNNList<DoubleDistance> knnlist = result.get(id);
        meansize.put(knnlist.size());
      }
      org.junit.Assert.assertEquals("Euclidean mean 2NN", mean2nnEuclid, meansize.getMean(), 0.00001);
      org.junit.Assert.assertEquals("Euclidean variance 2NN", var2nnEuclid, meansize.getSampleVariance(), 0.00001);
    }
    // Manhattan
    {
      KNNJoin<DoubleVector, DoubleDistance, ?, ?> knnjoin = new KNNJoin<DoubleVector, DoubleDistance, RStarTreeNode, SpatialEntry>(ManhattanDistanceFunction.STATIC, 2);
      DataStore<KNNList<DoubleDistance>> result = knnjoin.run(db);


      MeanVariance meansize = new MeanVariance();
      for(DBID id : relation.getDBIDs()) {
        KNNList<DoubleDistance> knnlist = result.get(id);
        meansize.put(knnlist.size());
      }
      org.junit.Assert.assertEquals("Manhattan mean 2NN", mean2nnManhattan, meansize.getMean(), 0.00001);
      org.junit.Assert.assertEquals("Manhattan variance 2NN", var2nnManhattan, meansize.getSampleVariance(), 0.00001);
    }
  }

View Full Code Here

Examples of de.lmu.ifi.dbs.elki.math.MeanVariance

  }


  @Override
  public void prepare(OutlierResult or) {
    if(usemean) {
      MeanVariance mv = new MeanVariance();
      DoubleMinMax mm = (max == null) ? new DoubleMinMax() : null;
      boolean skippedzeros = false;
      for(DBID id : or.getScores().iterDBIDs()) {
        double val = or.getScores().get(id);
        if(nozeros && val == 0.0) {
          skippedzeros = true;
          continue;
        }
        if(!Double.isNaN(val) && !Double.isInfinite(val)) {
          mv.put(val);
        }
        if(max == null) {
          mm.put(val);
        }
      }
      if(skippedzeros && mm.getMin() == mm.getMax()) {
        mm.put(0.0);
        mv.put(0.0);
      }
      min = mv.getMean();
      if(max == null) {
        max = mm.getMax();
      }
    }
    else {

View Full Code Here

Examples of de.lmu.ifi.dbs.elki.math.MeanVariance

   */
  double scaledmax;


  @Override
  public void prepare(OutlierResult or) {
    MeanVariance mv = new MeanVariance();
    DoubleMinMax minmax = new DoubleMinMax();


    for(DBID id : or.getScores().iterDBIDs()) {
      double val = or.getScores().get(id);
      if(!Double.isNaN(val) && !Double.isInfinite(val)) {
        mv.put(val);
        minmax.put(val);
      }
    }


    mean = mv.getMean();
    stddev = mv.getSampleStddev();
    scaledmax = getScaled(minmax.getMax());
    scaledmin = getScaled(minmax.getMin());
  }

View Full Code Here

Examples of de.lmu.ifi.dbs.elki.math.MeanVariance

  }


  @Override
  public void prepare(OutlierResult or) {
    meta = or.getOutlierMeta();
    MeanVariance mv = new MeanVariance();
    for(DBID id : or.getScores().iterDBIDs()) {
      double score = or.getScores().get(id);
      score = preScale(score);
      if(!Double.isNaN(score) && !Double.isInfinite(score)) {
        mv.put(score);
      }
    }
    final double mean = mv.getMean();
    final double var = mv.getSampleVariance();
    k = (mean * mean) / var;
    theta = var / mean;
    atmean = GammaDistribution.regularizedGammaP(k, mean / theta);
    // logger.warning("Mean:"+mean+" Var:"+var+" Theta: "+theta+" k: "+k+" valatmean"+atmean);
  }

View Full Code Here

Examples of de.lmu.ifi.dbs.elki.math.MeanVariance

  }


  @Override
  public void prepare(OutlierResult or) {
    // Initial parameters - are these defaults sounds?
    MeanVariance mv = new MeanVariance();
    for(DBID id : or.getScores().iterDBIDs()) {
      double val = or.getScores().get(id);
      if(!Double.isNaN(val) && !Double.isInfinite(val)) {
        mv.put(val);
      }
    }
    double curMu = mv.getMean() * 2;
    if(curMu == 0) {
      curMu = Double.MIN_NORMAL;
    }
    double curSigma = Math.max(mv.getSampleStddev(), Double.MIN_NORMAL);
    double curLambda = Math.min(1.0 / curMu, Double.MAX_VALUE);
    double curAlpha = 0.05;


    ArrayDBIDs ids = DBIDUtil.ensureArray(or.getScores().getDBIDs());
    // TODO: stop condition!

View Full Code Here

Examples of de.lmu.ifi.dbs.elki.math.MeanVariance

      }
    }
    max = mm.getMax();
    mlogmax = -Math.log(mm.getMin() / max);
    // with the prescaling, do Gamma Scaling.
    MeanVariance mv = new MeanVariance();
    for(DBID id : or.getScores().iterDBIDs()) {
      double score = or.getScores().get(id);
      score = preScale(score);
      if(!Double.isNaN(score) && !Double.isInfinite(score)) {
        mv.put(score);
      }
    }
    final double mean = mv.getMean();
    final double var = mv.getSampleVariance();
    k = (mean * mean) / var;
    theta = var / mean;
    atmean = GammaDistribution.regularizedGammaP(k, mean / theta);
    // logger.warning("Mean:"+mean+" Var:"+var+" Theta: "+theta+" k: "+k+" valatmean"+atmean);
  }

View Full Code Here

0 1 2 3 4 5

TOP

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.