001/**
002 * Copyright (c) 2011, The University of Southampton and the individual contributors.
003 * All rights reserved.
004 *
005 * Redistribution and use in source and binary forms, with or without modification,
006 * are permitted provided that the following conditions are met:
007 *
008 *   *  Redistributions of source code must retain the above copyright notice,
009 *      this list of conditions and the following disclaimer.
010 *
011 *   *  Redistributions in binary form must reproduce the above copyright notice,
012 *      this list of conditions and the following disclaimer in the documentation
013 *      and/or other materials provided with the distribution.
014 *
015 *   *  Neither the name of the University of Southampton nor the names of its
016 *      contributors may be used to endorse or promote products derived from this
017 *      software without specific prior written permission.
018 *
019 * THIS SOFTWARE IS PROVIDED BY THE COPYRIGHT HOLDERS AND CONTRIBUTORS "AS IS" AND
020 * ANY EXPRESS OR IMPLIED WARRANTIES, INCLUDING, BUT NOT LIMITED TO, THE IMPLIED
021 * WARRANTIES OF MERCHANTABILITY AND FITNESS FOR A PARTICULAR PURPOSE ARE
022 * DISCLAIMED. IN NO EVENT SHALL THE COPYRIGHT OWNER OR CONTRIBUTORS BE LIABLE FOR
023 * ANY DIRECT, INDIRECT, INCIDENTAL, SPECIAL, EXEMPLARY, OR CONSEQUENTIAL DAMAGES
024 * (INCLUDING, BUT NOT LIMITED TO, PROCUREMENT OF SUBSTITUTE GOODS OR SERVICES;
025 * LOSS OF USE, DATA, OR PROFITS; OR BUSINESS INTERRUPTION) HOWEVER CAUSED AND ON
026 * ANY THEORY OF LIABILITY, WHETHER IN CONTRACT, STRICT LIABILITY, OR TORT
027 * (INCLUDING NEGLIGENCE OR OTHERWISE) ARISING IN ANY WAY OUT OF THE USE OF THIS
028 * SOFTWARE, EVEN IF ADVISED OF THE POSSIBILITY OF SUCH DAMAGE.
029 */
030package org.openimaj.ml.linear.learner.matlib.loss;
031
032import org.apache.logging.log4j.Logger;
033import org.apache.logging.log4j.LogManager;
034
035import org.openimaj.math.matrix.MatlibMatrixUtils;
036
037import ch.akuhn.matrix.Matrix;
038import ch.akuhn.matrix.Vector;
039
040
041public class MatSquareLossFunction extends LossFunction{
042        Logger logger = LogManager.getLogger(MatSquareLossFunction.class);
043        public MatSquareLossFunction() {
044        }
045        @Override
046        public Matrix gradient(Matrix W) {
047                Matrix ret = W.newInstance();
048                Matrix resid = MatlibMatrixUtils.dotProduct(X, W);
049                if(this.bias!=null)
050                {
051                        MatlibMatrixUtils.plusInplace(resid, this.bias);
052                }
053                MatlibMatrixUtils.minusInplace(resid, Y);
054                for (int t = 0; t < resid.columnCount(); t++) {
055                        Vector row = this.X.row(t);
056                        row.times(resid.get(t, t));
057                        MatlibMatrixUtils.setSubMatrixCol(ret, 0, t, row);
058                }
059                return ret;
060        }
061        @Override
062        public double eval(Matrix W) {
063                Matrix resid = null;
064                if(W == null){
065                        resid = X;
066                } else {
067                        resid = MatlibMatrixUtils.dotProduct(X,W);
068                }
069                Matrix vnobias = MatlibMatrixUtils.copy(X);
070                if(this.bias!=null)
071                {
072                        MatlibMatrixUtils.plusInplace(resid, bias);
073                }
074                Matrix v =  MatlibMatrixUtils.copy(resid);
075                MatlibMatrixUtils.minusInplace(resid,Y);
076                double retval = 0;
077                
078                for (int t = 0; t < resid.columnCount(); t++) {
079                        double loss = resid.get(t, t);
080                        retval += loss * loss;
081                        logger.debug(
082                                        String.format(
083                                                        "yr=%d,y=%3.2f,v=%3.2f,v(no bias)=%2.5f,error=%2.5f,serror=%2.5f",
084                                                        t, Y.get(t, t), v.get(t, t), vnobias.get(t,t), loss, loss*loss
085                                                        )
086                                        );
087                }
088                return retval;
089        }
090        @Override
091        public boolean isMatrixLoss() {
092                return true;
093        }
094}