/**
* Copyright (c) 2009/09-2012/08, Regents of the University of Colorado
* All rights reserved.
*
* Redistribution and use in source and binary forms, with or without
* modification, are permitted provided that the following conditions are met:
*
* 1. Redistributions of source code must retain the above copyright notice, this
* list of conditions and the following disclaimer.
* 2. 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.
*
* 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.
*/
/**
* Copyright 2012/09-2013/04, 2013/11-Present, University of Massachusetts Amherst
* Copyright 2013/05-2013/10, IPSoft Inc.
*
* Licensed under the Apache License, Version 2.0 (the "License");
* you may not use this file except in compliance with the License.
* You may obtain a copy of the License at
*
* http://www.apache.org/licenses/LICENSE-2.0
*
* Unless required by applicable law or agreed to in writing, software
* distributed under the License is distributed on an "AS IS" BASIS,
* WITHOUT WARRANTIES OR CONDITIONS OF ANY KIND, either express or implied.
* See the License for the specific language governing permissions and
* limitations under the License.
*/
package com.clearnlp.classification.algorithm.old;
import java.util.ArrayList;
import java.util.Arrays;
import java.util.Random;
import com.carrotsearch.hppc.IntArrayList;
import com.clearnlp.classification.model.AbstractModel;
import com.clearnlp.classification.train.AbstractTrainSpace;
import com.clearnlp.util.UTArray;
import com.clearnlp.util.UTMath;
/**
* Abstract algorithm.
* @since 1.3.2
* @author Jinho D. Choi ({@code jdchoi77@gmail.com})
*/
abstract public class AbstractAdaGrad extends AbstractMulticlass
{
protected final int MAX_ITER = 1000;
protected double d_alpha;
protected double d_rho;
protected double d_eps;
public AbstractAdaGrad(double alpha, double rho, double eps)
{
d_alpha = alpha;
d_rho = rho;
d_eps = eps;
}
abstract protected boolean update(int L, int y, int[] x, double[] v, double[] gs, double[] weights);
abstract protected boolean update(int L, int y, int[] x, double[] v, double[] gs, double[] cWeights, double[] aWeights, int count);
@Override
public void updateWeights(AbstractTrainSpace space, boolean average)
{
final int D = space.getFeatureSize();
final int L = space.getLabelSize();
final int N = space.getInstanceSize();
final int WS = D * L;
IntArrayList ys = space.getYs();
ArrayList<int[]> xs = space.getXs();
ArrayList<double[]> vs = space.getVs();
AbstractModel model = space.getModel();
double[] cWeights = new double[WS];
double[] aWeights = average ? new double[WS] : null;
double[] gs = new double[WS];
double stdev, prevScore, currScore = 0;
int[] indices = UTArray.range(N);
int i, j, correct, count = 1;
int yi;
int[] xi;
double[] vi = null;
for (i=0; i<MAX_ITER; i++)
{
UTArray.shuffle(new Random(5), indices, N);
prevScore = currScore;
Arrays.fill(gs, 0);
correct = 0;
for (j=0; j<N; j++)
{
yi = ys.get(indices[j]);
xi = xs.get(indices[j]);
if (space.hasWeight()) vi = vs.get(indices[j]);
if (average)
{
if (!update(L, yi, xi, vi, gs, cWeights, aWeights, count))
correct++;
count++;
}
else if (!update(L, yi, xi, vi, gs, cWeights))
correct++;
}
currScore = 100d * correct / N;
stdev = UTMath.stdev(prevScore, currScore);
LOG.info(String.format("%4d: acc = %5.2f, stdev = %7.4f\n", i+1, currScore, stdev));
if (stdev < d_eps) break;
}
if (average) model.setWeights(getWeights(cWeights, aWeights, count));
else model.setWeights(UTArray.toFloatArray(cWeights));
}
protected double getCost(int L, double[] gs, int y, int x)
{
return d_alpha / (d_rho + Math.sqrt(gs[getWeightIndex(L, y, x)]));
}
protected float[] getWeights(double[] cWeights, double[] aWeights, int count)
{
int i, size = cWeights.length;
float[] fs = new float[size];
double c = 1d / count;
for (i=0; i<size; i++)
fs[i] = (float)(cWeights[i] - c * aWeights[i]);
return fs;
}
protected void updateWeightForAveraging(int idx, double cost, double[] cWeights, double[] aWeights, int count)
{
cWeights[idx] += cost;
aWeights[idx] += cost * count;
}
}