/**
 * Copyright (c) 2011, The University of Southampton and the individual contributors.
 * All rights reserved.
 *
 * Redistribution and use in source and binary forms, with or without modification,
 * are permitted provided that the following conditions are met:
 *
 *   *  Redistributions of source code must retain the above copyright notice,
 *      this list of conditions and the following disclaimer.
 *
 *   *  Redistributions in binary form must reproduce the above copyright notice,
 *      this list of conditions and the following disclaimer in the documentation
 *      and/or other materials provided with the distribution.
 *
 *   *  Neither the name of the University of Southampton nor the names of its
 *      contributors may be used to endorse or promote products derived from this
 *      software without specific prior written permission.
 *
 * THIS SOFTWARE IS PROVIDED BY THE COPYRIGHT HOLDERS AND CONTRIBUTORS "AS IS" AND
 * ANY EXPRESS OR IMPLIED WARRANTIES, INCLUDING, BUT NOT LIMITED TO, THE IMPLIED
 * WARRANTIES OF MERCHANTABILITY AND FITNESS FOR A PARTICULAR PURPOSE ARE
 * DISCLAIMED. IN NO EVENT SHALL THE COPYRIGHT OWNER OR CONTRIBUTORS BE LIABLE FOR
 * ANY DIRECT, INDIRECT, INCIDENTAL, SPECIAL, EXEMPLARY, OR CONSEQUENTIAL DAMAGES
 * (INCLUDING, BUT NOT LIMITED TO, PROCUREMENT OF SUBSTITUTE GOODS OR SERVICES;
 * LOSS OF USE, DATA, OR PROFITS; OR BUSINESS INTERRUPTION) HOWEVER CAUSED AND ON
 * ANY THEORY OF LIABILITY, WHETHER IN CONTRACT, STRICT LIABILITY, OR TORT
 * (INCLUDING NEGLIGENCE OR OTHERWISE) ARISING IN ANY WAY OUT OF THE USE OF THIS
 * SOFTWARE, EVEN IF ADVISED OF THE POSSIBILITY OF SUCH DAMAGE.
 */
package org.openimaj.math.statistics.distribution;

import gnu.trove.procedure.TObjectDoubleProcedure;

import java.util.ArrayList;
import java.util.List;
import java.util.Random;

import org.openimaj.math.statistics.distribution.kernel.UnivariateKernel;
import org.openimaj.util.pair.ObjectDoublePair;
import org.openimaj.util.tree.DoubleKDTree;

/**
 * A Parzen window kernel density estimate using a univariate kernel and
 * Euclidean distance. Uses a KD-Tree to for efficient neighbour search.
 * 
 * @author Jonathon Hare (jsh2@ecs.soton.ac.uk)
 * 
 */
public class MultivariateKernelDensityEstimate extends AbstractMultivariateDistribution {
        double[][] data;
        UnivariateKernel kernel;
        private double bandwidth;
        DoubleKDTree tree;

        /**
         * Construct with the given data, kernel and bandwidth
         * 
         * @param data
         *            the data
         * @param kernel
         *            the kernel
         * @param bandwidth
         *            the bandwidth
         */
        public MultivariateKernelDensityEstimate(double[][] data, UnivariateKernel kernel, double bandwidth) {
                this.data = data;
                this.tree = new DoubleKDTree(data);
                this.kernel = kernel;
                this.bandwidth = bandwidth;
        }

        /**
         * Construct with the given data, kernel and bandwidth
         * 
         * @param data
         *            the data
         * @param kernel
         *            the kernel
         * @param bandwidth
         *            the bandwidth
         */
        public MultivariateKernelDensityEstimate(List<double[]> data, UnivariateKernel kernel, double bandwidth)
        {
                this.data = data.toArray(new double[data.size()][]);
                this.tree = new DoubleKDTree(this.data);
                this.kernel = kernel;
                this.bandwidth = bandwidth;
        }

        @Override
        public double[] sample(Random rng) {
                final double[] pt = data[rng.nextInt(data.length)].clone();

                for (int i = 0; i < pt.length; i++) {
                        pt[i] = pt[i] + kernel.sample(rng) * this.getBandwidth();
                }

                return pt;
        }

        @Override
        public double estimateProbability(double[] sample) {
                final double[] prob = new double[1];
                final int[] count = new int[1];

                tree.coordinateRadiusSearch(sample, kernel.getCutOff() * getBandwidth(), new TObjectDoubleProcedure<double[]>() {
                        @Override
                        public boolean execute(double[] point, double distance) {
                                prob[0] += kernel.evaluate(Math.sqrt(distance) / getBandwidth());
                                count[0]++;

                                return true;
                        }
                });

                return prob[0] / (getBandwidth() * count[0]);
        }

        /**
         * Get the underlying points that support the KDE within the window around
         * the given point. Each point is returned together with its own density
         * estimate.
         * 
         * @param sample
         *            the point in the centre of the window
         * @return the points in the window
         */
        public List<ObjectDoublePair<double[]>> getSupport(double[] sample) {
                final List<ObjectDoublePair<double[]>> support = new ArrayList<ObjectDoublePair<double[]>>();

                tree.coordinateRadiusSearch(sample, kernel.getCutOff() * getBandwidth(), new TObjectDoubleProcedure<double[]>() {
                        @Override
                        public boolean execute(double[] a, double b) {
                                support.add(ObjectDoublePair.pair(a, kernel.evaluate(Math.sqrt(b) / getBandwidth())));

                                return true;
                        }
                });

                return support;
        }

        /**
         * Get the underlying data
         * 
         * @return the data
         */
        public double[][] getData() {
                return data;
        }

        /**
         * Get the bandwidth
         * 
         * @return the bandwidth
         */
        public double getBandwidth() {
                return bandwidth;
        }

        /**
         * Get the bandwidth scaled by the kernel support.
         * 
         * @see UnivariateKernel#getCutOff()
         * 
         * @return the scaled bandwidth
         */
        public double getScaledBandwidth() {
                return bandwidth * this.kernel.getCutOff();
        }

        @Override
        public double[] estimateLogProbability(double[][] x) {
                final double[] lps = new double[x.length];
                for (int i = 0; i < x.length; i++)
                        lps[i] = estimateLogProbability(x[i]);
                return lps;
        }
}