Examples with DescriptiveStatistics org.apache.commons.math3.stat.descriptive.DescriptiveStatistics used on opensource projects

Example 6 with DescriptiveStatistics

use of org.apache.commons.math3.stat.descriptive.DescriptiveStatistics in project GDSC-SMLM by aherbert.

the class PSFCreator method getLimits.

/**
	 * Get the limits of the array ignoring outliers more than 1.5x the inter quartile range
	 * 
	 * @param data
	 * @return
	 */
private double[] getLimits(double[] data) {
    double[] limits = Maths.limits(data);
    DescriptiveStatistics stats = new DescriptiveStatistics(data);
    double lower = stats.getPercentile(25);
    double upper = stats.getPercentile(75);
    double iqr = (upper - lower) * 2;
    limits[0] = FastMath.max(lower - iqr, limits[0]);
    limits[1] = FastMath.min(upper + iqr, limits[1]);
    return limits;
}

Example 7 with DescriptiveStatistics

use of org.apache.commons.math3.stat.descriptive.DescriptiveStatistics in project lucene-solr by apache.

the class DescribeEvaluator method evaluate.

public Tuple evaluate(Tuple tuple) throws IOException {
    if (subEvaluators.size() != 1) {
        throw new IOException("describe expects 1 column as a parameters");
    }
    StreamEvaluator colEval = subEvaluators.get(0);
    List<Number> numbers = (List<Number>) colEval.evaluate(tuple);
    DescriptiveStatistics descriptiveStatistics = new DescriptiveStatistics();
    for (Number n : numbers) {
        descriptiveStatistics.addValue(n.doubleValue());
    }
    Map map = new HashMap();
    map.put("max", descriptiveStatistics.getMax());
    map.put("mean", descriptiveStatistics.getMean());
    map.put("min", descriptiveStatistics.getMin());
    map.put("stdev", descriptiveStatistics.getStandardDeviation());
    map.put("sum", descriptiveStatistics.getSum());
    map.put("N", descriptiveStatistics.getN());
    map.put("var", descriptiveStatistics.getVariance());
    map.put("kurtosis", descriptiveStatistics.getKurtosis());
    map.put("skewness", descriptiveStatistics.getSkewness());
    map.put("popVar", descriptiveStatistics.getPopulationVariance());
    map.put("geometricMean", descriptiveStatistics.getGeometricMean());
    map.put("sumsq", descriptiveStatistics.getSumsq());
    return new Tuple(map);
}

Example 8 with DescriptiveStatistics

use of org.apache.commons.math3.stat.descriptive.DescriptiveStatistics in project webofneeds by researchstudio-sat.

the class Kneedle method normalize.

private double[] normalize(final double[] values) {
    double[] normalized = new double[values.length];
    DescriptiveStatistics stats = new DescriptiveStatistics(values);
    for (int i = 0; i < values.length; i++) {
        normalized[i] = (values[i] - stats.getMin()) / (stats.getMax() - stats.getMin());
    }
    return normalized;
}

Example 9 with DescriptiveStatistics

use of org.apache.commons.math3.stat.descriptive.DescriptiveStatistics in project webofneeds by researchstudio-sat.

the class Kneedle method detectKneeOrElbowPoints.

/**
 * Detect all knee points in a curve according to "Kneedle" algorithm.
 * Alternatively this method can detect elbow points instead of knee points.
 *
 * @param x x-coordintes of curve, must be increasing in value
 * @param y y-coordinates of curve
 * @param detectElbows if true detects elbow points, if false detects knee points
 * @return array of indices of knee (or elbow) points in the curve
 */
private int[] detectKneeOrElbowPoints(final double[] x, final double[] y, boolean detectElbows) {
    checkConstraints(x, y);
    List<Integer> kneeIndices = new LinkedList<Integer>();
    List<Integer> lmxIndices = new LinkedList<Integer>();
    List<Double> lmxThresholds = new LinkedList<Double>();
    double[] xn = normalize(x);
    double[] yn = normalize(y);
    // compute the y difference values
    double[] yDiff = new double[y.length];
    for (int i = 0; i < y.length; i++) {
        yDiff[i] = yn[i] - xn[i];
    }
    // original yDiff curve
    if (detectElbows) {
        DescriptiveStatistics stats = new DescriptiveStatistics(yDiff);
        for (int i = 0; i < yDiff.length; i++) {
            yDiff[i] = stats.getMax() - yDiff[i];
        }
    }
    // find local maxima, compute threshold values and detect knee points
    boolean detectKneeForLastLmx = false;
    for (int i = 1; i < y.length - 1; i++) {
        // than for its left and right neighbour => local maximum
        if (yDiff[i] > yDiff[i - 1] && yDiff[i] > yDiff[i + 1]) {
            // local maximum found
            lmxIndices.add(i);
            // compute the threshold value for this local maximum
            // NOTE: As stated in the paper the threshold Tlmx is computed. Since the mean distance of all consecutive
            // x-values summed together for a normalized function is always (1 / (n -1)) we do not have to compute the
            // whole sum here as stated in the paper.
            double tlmx = yDiff[i] - sensitivity / (xn.length - 1);
            lmxThresholds.add(tlmx);
            // try to find out if the current local maximum is a knee point
            detectKneeForLastLmx = true;
        }
        // check for new knee point
        if (detectKneeForLastLmx) {
            if (yDiff[i + 1] < lmxThresholds.get(lmxThresholds.size() - 1)) {
                // knee detected
                kneeIndices.add(lmxIndices.get(lmxIndices.size() - 1));
                detectKneeForLastLmx = false;
            }
        }
    }
    int[] knees = new int[kneeIndices.size()];
    for (int i = 0; i < kneeIndices.size(); i++) {
        knees[i] = kneeIndices.get(i);
    }
    return knees;
}

Example 10 with DescriptiveStatistics

use of org.apache.commons.math3.stat.descriptive.DescriptiveStatistics in project cloudsim-plus by manoelcampos.

the class MathUtil method getStatistics.

/**
 * Gets an object to compute descriptive statistics for an list of numbers.
 *
 * @param list the list of numbers. Must not be null.
 * @return descriptive statistics for the list of numbers.
 */
public static DescriptiveStatistics getStatistics(final List<Double> list) {
    final DescriptiveStatistics stats = new DescriptiveStatistics();
    list.forEach(stats::addValue);
    return stats;
}