edu.byu.nlp.math.CachedDifferentiableMultivariateFunction.java Source code

Introduction

Here is the source code for edu.byu.nlp.math.CachedDifferentiableMultivariateFunction.java

Source

/**
 * Copyright 2011 Brigham Young University
 *
 * 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 edu.byu.nlp.math;

import org.apache.commons.math3.analysis.DifferentiableMultivariateFunction;
import org.apache.commons.math3.analysis.MultivariateFunction;
import org.apache.commons.math3.analysis.MultivariateVectorFunction;

import edu.byu.nlp.util.Caches;
import edu.byu.nlp.util.ValueSupplier;

/**
 * A {@link MultivariateRealFunction} that computes the value and gradient at the same time and caches the result
 * between invocations of {@code value()} and {@code gradient()}. Useful when there is overlap in the computation of the
 * value and gradient that is non-trivial computationally.
 * 
 * @author rah67
 *
 */
public class CachedDifferentiableMultivariateFunction implements DifferentiableMultivariateFunction {

    public static class ValueAndGradient {
        private final double value;
        private final double[] gradient;

        public ValueAndGradient(double value, double[] gradient) {
            this.value = value;
            this.gradient = gradient;
        }

        public double getValue() {
            return value;
        }

        public double[] getGradient() {
            return gradient;
        }
    }

    private class Gradient implements MultivariateVectorFunction {
        @Override
        public double[] value(double[] x) {
            return cachedValueSupplier.get(x).getGradient();
        }
    }

    private final ValueSupplier<double[], ValueAndGradient> cachedValueSupplier;
    private final MultivariateVectorFunction gradient;

    /**
     * @param valueSupplier supplies values on a cache miss.
     */
    public CachedDifferentiableMultivariateFunction(ValueSupplier<double[], ValueAndGradient> valueSupplier) {
        cachedValueSupplier = Caches.lastValueReferenceCache(valueSupplier);
        gradient = new Gradient();
    }

    /**
     * {@inheritDoc}
     */
    @Override
    public double value(double[] x) {
        return cachedValueSupplier.get(x).getValue();
    }

    /**
     * {@inheritDoc}
     */
    @Override
    public MultivariateVectorFunction gradient() {
        return gradient;
    }

    /**
     * {@inheritDoc}
     */
    @Override
    public MultivariateFunction partialDerivative(final int i) {
        return new MultivariateFunction() {
            @Override
            public double value(double[] x) {
                return cachedValueSupplier.get(x).getGradient()[i];
            }
        };
    }
}