Examples with MultivariateDistributionLikelihood dr.inference.distribution.MultivariateDistributionLikelihood used on opensource projects

Example 6 with MultivariateDistributionLikelihood

use of dr.inference.distribution.MultivariateDistributionLikelihood in project beast-mcmc by beast-dev.

the class GradientWrapperParser method parseXMLObject.

@Override
public Object parseXMLObject(XMLObject xo) throws XMLParseException {
    Object obj = xo.getChild(0);
    if (obj instanceof MultivariateDistributionLikelihood) {
        final MultivariateDistributionLikelihood mdl = (MultivariateDistributionLikelihood) xo.getChild(MultivariateDistributionLikelihood.class);
        if (!(mdl.getDistribution() instanceof GradientProvider)) {
            throw new XMLParseException("Not a gradient provider");
        }
        final GradientProvider provider = (GradientProvider) mdl.getDistribution();
        final Parameter parameter = mdl.getDataParameter();
        return new GradientWrtParameterProvider.ParameterWrapper(provider, parameter, mdl);
    } else if (obj instanceof EmpiricalDistributionLikelihood) {
        final Parameter parameter = (Parameter) xo.getChild(Parameter.class);
        return new GradientWrtParameterProvider.ParameterWrapper((GradientProvider) obj, parameter, (Likelihood) obj);
    } else {
        DistributionLikelihood dl = (DistributionLikelihood) obj;
        if (!(dl.getDistribution() instanceof GradientProvider)) {
            throw new XMLParseException("Not a gradient provider");
        }
        final GradientProvider provider = (GradientProvider) dl.getDistribution();
        final Parameter parameter = (Parameter) xo.getChild(Parameter.class);
        return new GradientWrtParameterProvider.ParameterWrapper(provider, parameter, dl);
    }
}

Example 7 with MultivariateDistributionLikelihood

use of dr.inference.distribution.MultivariateDistributionLikelihood in project beast-mcmc by beast-dev.

the class EllipticalSliceOperatorParser method parseXMLObject.

public Object parseXMLObject(XMLObject xo) throws XMLParseException {
    final double weight = xo.getDoubleAttribute(MCMCOperator.WEIGHT);
    final Parameter variable = (Parameter) xo.getChild(Parameter.class);
    boolean drawByRowTemp = false;
    if (xo.hasAttribute(DRAW_BY_ROW)) {
        drawByRowTemp = xo.getBooleanAttribute(DRAW_BY_ROW);
    }
    final boolean drawByRow = drawByRowTemp;
    boolean signal = xo.getAttribute(SIGNAL_CONSTITUENT_PARAMETERS, true);
    if (!signal) {
        Parameter possiblyCompound = variable;
        if (variable instanceof MaskedParameter) {
            possiblyCompound = ((MaskedParameter) variable).getUnmaskedParameter();
        }
        if (!(possiblyCompound instanceof CompoundParameter)) {
            signal = true;
        }
    }
    double bracketAngle = xo.getAttribute(BRACKET_ANGLE, 0.0);
    boolean translationInvariant = xo.getAttribute(TRANSLATION_INVARIANT, false);
    boolean rotationInvariant = xo.getAttribute(ROTATION_INVARIANT, false);
    GaussianProcessRandomGenerator gaussianProcess = (GaussianProcessRandomGenerator) xo.getChild(GaussianProcessRandomGenerator.class);
    if (gaussianProcess == null) {
        final MultivariateDistributionLikelihood likelihood = (MultivariateDistributionLikelihood) xo.getChild(MultivariateDistributionLikelihood.class);
        if (!(likelihood.getDistribution() instanceof GaussianProcessRandomGenerator)) {
            throw new XMLParseException("Elliptical slice sampling only works for multivariate normally distributed random variables");
        }
        if (likelihood.getDistribution() instanceof MultivariateNormalDistribution)
            gaussianProcess = (MultivariateNormalDistribution) likelihood.getDistribution();
        if (likelihood.getDistribution() instanceof MultivariateNormalDistributionModel)
            gaussianProcess = (MultivariateNormalDistributionModel) likelihood.getDistribution();
    }
    EllipticalSliceOperator operator = new EllipticalSliceOperator(variable, gaussianProcess, drawByRow, signal, bracketAngle, translationInvariant, rotationInvariant);
    operator.setWeight(weight);
    return operator;
}

Example 8 with MultivariateDistributionLikelihood

use of dr.inference.distribution.MultivariateDistributionLikelihood in project beast-mcmc by beast-dev.

the class MultivariateNormalGibbsOperatorParser method parseXMLObject.

@Override
public Object parseXMLObject(XMLObject xo) throws XMLParseException {
    MultivariateDistributionLikelihood prior = (MultivariateDistributionLikelihood) xo.getChild(PRIOR).getChild(MultivariateDistributionLikelihood.class);
    MultivariateDistributionLikelihood likelihood = (MultivariateDistributionLikelihood) xo.getChild(LIKELIHOOD).getChild(MultivariateDistributionLikelihood.class);
    // CompoundParameter data = (CompoundParameter) xo.getChild(CompoundParameter.class);
    String weightTemp = (String) xo.getAttribute(WEIGHT);
    Double weight = Double.parseDouble(weightTemp);
    try {
        // To change body of implemented methods use File | Settings | File Templates.
        return new MultivariateNormalGibbsOperator(likelihood, prior, weight);
    } catch (IllegalDimension illegalDimension) {
        // To change body of catch statement use File | Settings | File Templates.
        illegalDimension.printStackTrace();
    }
    return null;
}

Example 9 with MultivariateDistributionLikelihood

use of dr.inference.distribution.MultivariateDistributionLikelihood in project beast-mcmc by beast-dev.

the class ManyUniformGeoDistributionModelParser method parseXMLObject.

public Object parseXMLObject(XMLObject xo) throws XMLParseException {
    List<GeoSpatialDistribution> distributions = new ArrayList<GeoSpatialDistribution>();
    List<Parameter> parameters = new ArrayList<Parameter>();
    List<Likelihood> oldLikelihood = new ArrayList<Likelihood>();
    for (int i = 0; i < xo.getChildCount(); ++i) {
        Object cxo = xo.getChild(i);
        MultivariateDistributionLikelihood ab = null;
        if (cxo instanceof MultivariateDistributionLikelihood) {
            ab = (MultivariateDistributionLikelihood) cxo;
        } else if (cxo instanceof CachedDistributionLikelihood) {
            ab = (MultivariateDistributionLikelihood) ((CachedDistributionLikelihood) cxo).getDistributionLikelihood();
        }
        if (ab != null) {
            parameters.add(ab.getDataParameter());
            MultivariateDistribution md = ab.getDistribution();
            oldLikelihood.add((Likelihood) cxo);
            if (md instanceof GeoSpatialDistribution) {
                distributions.add((GeoSpatialDistribution) md);
            } else {
                throw new XMLParseException("Unhandled distribution type in '" + xo.getId() + "'");
            }
        }
    }
    return new ManyUniformGeoDistributionModel(xo.getId(), parameters, distributions, oldLikelihood);
}

Example 10 with MultivariateDistributionLikelihood

use of dr.inference.distribution.MultivariateDistributionLikelihood in project beast-mcmc by beast-dev.

the class Polygon2DSampling method main.

// private boolean hasFillValue;
public static void main(String[] args) {
    List<AbstractPolygon2D> polygons = Polygon2D.readKMLFile(args[0]);
    System.out.println("Reading from file: " + args[0] + "\n");
    List<GeoSpatialDistribution> geoSpatialDistributions = new ArrayList<GeoSpatialDistribution>();
    // check if all the polygons have either a fillValue or do not have this fillValue
    boolean overAllFill = false;
    AbstractPolygon2D first = polygons.get(0);
    System.out.println("Polygon 0: " + first.getFillValue());
    boolean equalFills = first.hasFillValue();
    if (polygons.size() > 1) {
        for (int i = 1; i < polygons.size(); i++) {
            if (!(equalFills == polygons.get(i).hasFillValue())) {
                throw new RuntimeException("Inconsistent fillValue attributes provided.");
            } else {
                System.out.println("Polygon " + i + ": " + polygons.get(i).getFillValue());
            }
        }
    }
    if (equalFills) {
        double sum = 0.0;
        for (AbstractPolygon2D region : polygons) {
            sum += region.getFillValue();
        }
        if (Math.abs(sum - 1.0) > 1E-12) {
            throw new RuntimeException("Fill values in " + args[0] + " do not sum to 1 : " + sum);
        } else {
            System.out.println("Provided sampling probabilities sum to 1.0");
        }
    }
    overAllFill = equalFills;
    for (AbstractPolygon2D region : polygons) {
        geoSpatialDistributions.add(new GeoSpatialDistribution("", region, true));
    }
    System.out.println("\nTesting sampling probabilities:");
    MultivariateDistributionLikelihood likelihood = new MultivariateDistributionLikelihood(new MultiRegionGeoSpatialDistribution("", geoSpatialDistributions, false, overAllFill));
    double[] coordinates = new double[2];
    coordinates[0] = 12.68379879;
    coordinates[1] = 108.12982178;
    Parameter parameter = new Parameter.Default(coordinates);
    likelihood.addData(parameter);
    System.out.println("logL = " + likelihood.getLogLikelihood());
    System.out.println("L = " + Math.exp(likelihood.getLogLikelihood()));
    parameter.setParameterValue(0, 13.68379879);
    System.out.println("logL = " + likelihood.getLogLikelihood());
    System.out.println("L = " + Math.exp(likelihood.getLogLikelihood()));
    parameter.setParameterValue(0, 12.68);
    System.out.println("logL = " + likelihood.getLogLikelihood());
    System.out.println("L = " + Math.exp(likelihood.getLogLikelihood()));
    parameter.setParameterValue(0, 13.03387928);
    parameter.setParameterValue(1, 108.51604462);
    System.out.println("logL = " + likelihood.getLogLikelihood());
    System.out.println("L = " + Math.exp(likelihood.getLogLikelihood()));
    parameter.setParameterValue(0, 13.01971245);
    parameter.setParameterValue(1, 108.56764221);
    System.out.println("logL = " + likelihood.getLogLikelihood());
    System.out.println("L = " + Math.exp(likelihood.getLogLikelihood()));
    parameter.setParameterValue(0, 13.018);
    System.out.println("logL = " + likelihood.getLogLikelihood());
    System.out.println("L = " + Math.exp(likelihood.getLogLikelihood()));
// from 8 to 14 and from 104 to 110
/*System.out.println("\n\n\nplot(0,0,xlim=c(9,14),ylim=c(105,110))");
        Random rand = new Random();
        for (int i = 0; i < 100000; i++) {
            parameter.setParameterValue(0, rand.nextDouble()*6+9);
            parameter.setParameterValue(1, rand.nextDouble()*6+105);
            if (!Double.isInfinite(likelihood.getLogLikelihood())) {
                double color = Math.exp(likelihood.getLogLikelihood());
                //System.out.println(color);
                String col = "black";
                if (Math.abs(color - 0.106508875739645) < 1E-5) {
                    col = "red";
                } else if (Math.abs(color - 0.029585798816568) < 1E-5) {
                    col = "blue";
                } else if (Math.abs(color - 0.0769230769230769) < 1E-5) {
                    col = "green";
                } else if (Math.abs(color - 0.0355029585798817) < 1E-5) {
                    col = "grey";
                } else if (Math.abs(color - 0.124260355029586) < 1E-5) {
                    col = "purple";
                } else if (Math.abs(color - 0.112426035502959) < 1E-5) {
                    col = "orange";
                } else if (Math.abs(color - 0.136094674556213) < 1E-5) {
                    col = "yellow";
                } else if (Math.abs(color - 0.0946745562130177) < 1E-5) {
                    col = "burlywood";
                } else if (Math.abs(color - 0.284023668639053) < 1E-5) {
                    col = "deepskyblue";
                }
                System.out.println("points(" + parameter.getParameterValue(0) + "," + parameter.getParameterValue(1) + ",col=\"" + col + "\",pch=20,bg=\""+ col +"\")");
            }
        }*/
}