Apache
/
commons-math
mirror of https://github.com/apache/commons-math.git
1
0
Fork 0
Code Issues Packages Projects Releases Wiki Activity

Use Evaluation instead of PointVectorValuePair 

Use Evaluation instead of PointVectorValuePair in the ConvergenceChecker. This
gives the checkers access to more information, such as the rms and covariances.
The change also simplified the optimizer implementations since they no longer
have to keep track of the current function value.

A method was added to LeastSquaresFactory to convert between the two types of
checkers and a method added to LeastSquaresBuilder so that it can accept either
type. I would have prefered to do this through method overloading, but
overloading doesn't play well with generics.

git-svn-id: https://svn.apache.org/repos/asf/commons/proper/math/trunk@1569353 13f79535-47bb-0310-9956-ffa450edef68
This commit is contained in:
Luc Maisonobe 2014-02-18 14:32:44 +00:00
parent 3e18e999c7
commit a7a380f934
9 changed files with 76 additions and 51 deletions
Show Stats Download Patch File Download Diff File Expand all files Collapse all files

17
src/main/java/org/apache/commons/math3/fitting/leastsquares/GaussNewtonOptimizer.java
View File

@ -28,7 +28,6 @@ import org.apache.commons.math3.linear.QRDecomposition;
import org.apache.commons.math3.linear.RealMatrix;
import org.apache.commons.math3.linear.SingularMatrixException;
import org.apache.commons.math3.optim.ConvergenceChecker;
import org.apache.commons.math3.optim.PointVectorValuePair;
import org.apache.commons.math3.util.Incrementor;
/**
@ -123,7 +122,7 @@ public class GaussNewtonOptimizer implements LeastSquaresOptimizer {
//create local evaluation and iteration counts
final Incrementor evaluationCounter = lsp.getEvaluationCounter();
final Incrementor iterationCounter = lsp.getIterationCounter();
final ConvergenceChecker<PointVectorValuePair> checker
final ConvergenceChecker<Evaluation> checker
= lsp.getConvergenceChecker();
// Computation will be useless without a checker (see "for-loop").
@ -137,25 +136,23 @@ public class GaussNewtonOptimizer implements LeastSquaresOptimizer {
final double[] currentPoint = lsp.getStart();
// iterate until convergence is reached
PointVectorValuePair current = null;
Evaluation current = null;
while (true) {
iterationCounter.incrementCount();
// evaluate the objective function and its jacobian
PointVectorValuePair previous = current;
Evaluation previous = current;
// Value of the objective function at "currentPoint".
evaluationCounter.incrementCount();
final Evaluation value = lsp.evaluate(currentPoint);
final double[] currentObjective = value.computeValue();
final double[] currentResiduals = value.computeResiduals();
final RealMatrix weightedJacobian = value.computeJacobian();
current = new PointVectorValuePair(currentPoint, currentObjective);
current = lsp.evaluate(currentPoint);
final double[] currentResiduals = current.computeResiduals();
final RealMatrix weightedJacobian = current.computeJacobian();
// Check convergence.
if (previous != null) {
if (checker.converged(iterationCounter.getCount(), previous, current)) {
return new OptimumImpl(
value,
current,
evaluationCounter.getCount(),
iterationCounter.getCount());
}

3
src/main/java/org/apache/commons/math3/fitting/leastsquares/LeastSquaresAdapter.java
View File

@ -1,7 +1,6 @@
package org.apache.commons.math3.fitting.leastsquares;
import org.apache.commons.math3.optim.ConvergenceChecker;
import org.apache.commons.math3.optim.PointVectorValuePair;
import org.apache.commons.math3.util.Incrementor;
/**
@ -54,7 +53,7 @@ public class LeastSquaresAdapter implements LeastSquaresProblem {
}
/** {@inheritDoc} */
public ConvergenceChecker<PointVectorValuePair> getConvergenceChecker() {
public ConvergenceChecker<Evaluation> getConvergenceChecker() {
return problem.getConvergenceChecker();
}
}

17
src/main/java/org/apache/commons/math3/fitting/leastsquares/LeastSquaresBuilder.java
View File

@ -2,6 +2,7 @@ package org.apache.commons.math3.fitting.leastsquares;
import org.apache.commons.math3.analysis.MultivariateMatrixFunction;
import org.apache.commons.math3.analysis.MultivariateVectorFunction;
import org.apache.commons.math3.fitting.leastsquares.LeastSquaresProblem.Evaluation;
import org.apache.commons.math3.linear.RealMatrix;
import org.apache.commons.math3.optim.ConvergenceChecker;
import org.apache.commons.math3.optim.PointVectorValuePair;
@ -19,7 +20,7 @@ public class LeastSquaresBuilder {
/** max iterations */
private int maxIterations;
/** convergence checker */
private ConvergenceChecker<PointVectorValuePair> checker;
private ConvergenceChecker<Evaluation> checker;
/** model function */
private MultivariateVectorFunction model;
/** Jacobian function */
@ -69,11 +70,23 @@ public class LeastSquaresBuilder {
* @param checker the convergence checker.
* @return this
*/
public LeastSquaresBuilder checker(final ConvergenceChecker<PointVectorValuePair> checker) {
public LeastSquaresBuilder checker(final ConvergenceChecker<Evaluation> checker) {
this.checker = checker;
return this;
}
/**
* Configure the convergence checker.
* <p/>
* This function is an overloaded version of {@link #checker(ConvergenceChecker)}.
*
* @param checker the convergence checker.
* @return this
*/
public LeastSquaresBuilder checkerPair(final ConvergenceChecker<PointVectorValuePair> checker) {
return this.checker(LeastSquaresFactory.evaluationChecker(checker));
}
/**
* Configure the model function.
*

36
src/main/java/org/apache/commons/math3/fitting/leastsquares/LeastSquaresFactory.java
View File

@ -2,6 +2,7 @@ package org.apache.commons.math3.fitting.leastsquares;
import org.apache.commons.math3.analysis.MultivariateMatrixFunction;
import org.apache.commons.math3.analysis.MultivariateVectorFunction;
import org.apache.commons.math3.fitting.leastsquares.LeastSquaresProblem.Evaluation;
import org.apache.commons.math3.linear.Array2DRowRealMatrix;
import org.apache.commons.math3.linear.ArrayRealVector;
import org.apache.commons.math3.linear.DiagonalMatrix;
@ -40,7 +41,7 @@ public class LeastSquaresFactory {
public static LeastSquaresProblem create(final MultivariateJacobianFunction model,
final double[] observed,
final double[] start,
final ConvergenceChecker<PointVectorValuePair> checker,
final ConvergenceChecker<Evaluation> checker,
final int maxEvaluations,
final int maxIterations) {
return new LeastSquaresProblemImpl(
@ -70,7 +71,7 @@ public class LeastSquaresFactory {
final MultivariateMatrixFunction jacobian,
final double[] observed,
final double[] start,
final ConvergenceChecker<PointVectorValuePair> checker,
final ConvergenceChecker<Evaluation> checker,
final int maxEvaluations,
final int maxIterations) {
return create(
@ -102,7 +103,7 @@ public class LeastSquaresFactory {
final double[] observed,
final double[] start,
final RealMatrix weight,
final ConvergenceChecker<PointVectorValuePair> checker,
final ConvergenceChecker<Evaluation> checker,
final int maxEvaluations,
final int maxIterations) {
return weightMatrix(
@ -174,6 +175,35 @@ public class LeastSquaresFactory {
};
}
/**
* View a convergence checker specified for a {@link PointVectorValuePair} as one
* specified for an {@link Evaluation}.
*
* @param checker the convergence checker to adapt.
* @return a convergence checker that delegates to {@code checker}.
*/
public static ConvergenceChecker<Evaluation> evaluationChecker(
final ConvergenceChecker<PointVectorValuePair> checker
) {
return new ConvergenceChecker<Evaluation>() {
public boolean converged(final int iteration,
final Evaluation previous,
final Evaluation current) {
return checker.converged(
iteration,
new PointVectorValuePair(
previous.getPoint(),
previous.computeValue(),
false),
new PointVectorValuePair(
current.getPoint(),
current.computeValue(),
false)
);
}
};
}
/**
* Computes the square-root of the weight matrix.
*

4
src/main/java/org/apache/commons/math3/fitting/leastsquares/LeastSquaresProblem.java
View File

@ -1,8 +1,8 @@
package org.apache.commons.math3.fitting.leastsquares;
import org.apache.commons.math3.exception.DimensionMismatchException;
import org.apache.commons.math3.fitting.leastsquares.LeastSquaresProblem.Evaluation;
import org.apache.commons.math3.linear.RealMatrix;
import org.apache.commons.math3.optim.PointVectorValuePair;
/**
* The data necessary to define a non-linear least squares problem. Includes the observed
@ -12,7 +12,7 @@ import org.apache.commons.math3.optim.PointVectorValuePair;
*
* @version $Id$
*/
public interface LeastSquaresProblem extends OptimizationProblem<PointVectorValuePair> {
public interface LeastSquaresProblem extends OptimizationProblem<Evaluation> {
/**
* Gets the initial guess.

6
src/main/java/org/apache/commons/math3/fitting/leastsquares/LeastSquaresProblemImpl.java
View File

@ -17,11 +17,11 @@
package org.apache.commons.math3.fitting.leastsquares;
import org.apache.commons.math3.exception.DimensionMismatchException;
import org.apache.commons.math3.fitting.leastsquares.LeastSquaresProblem.Evaluation;
import org.apache.commons.math3.linear.RealMatrix;
import org.apache.commons.math3.linear.RealVector;
import org.apache.commons.math3.optim.AbstractOptimizationProblem;
import org.apache.commons.math3.optim.ConvergenceChecker;
import org.apache.commons.math3.optim.PointVectorValuePair;
import org.apache.commons.math3.util.Pair;
/**
@ -32,7 +32,7 @@ import org.apache.commons.math3.util.Pair;
* @since 3.3
*/
class LeastSquaresProblemImpl
extends AbstractOptimizationProblem<PointVectorValuePair>
extends AbstractOptimizationProblem<Evaluation>
implements LeastSquaresProblem {
/** Target values for the model function at optimum. */
@ -45,7 +45,7 @@ class LeastSquaresProblemImpl
LeastSquaresProblemImpl(final MultivariateJacobianFunction model,
final double[] target,
final double[] start,
final ConvergenceChecker<PointVectorValuePair> checker,
final ConvergenceChecker<Evaluation> checker,
final int maxEvaluations,
final int maxIterations) {
super(maxEvaluations, maxIterations, checker);

40
src/main/java/org/apache/commons/math3/fitting/leastsquares/LevenbergMarquardtOptimizer.java
View File

@ -23,7 +23,6 @@ import org.apache.commons.math3.linear.RealMatrix;
import org.apache.commons.math3.exception.ConvergenceException;
import org.apache.commons.math3.exception.util.LocalizedFormats;
import org.apache.commons.math3.optim.ConvergenceChecker;
import org.apache.commons.math3.optim.PointVectorValuePair;
import org.apache.commons.math3.util.Incrementor;
import org.apache.commons.math3.util.Precision;
import org.apache.commons.math3.util.FastMath;
@ -303,7 +302,7 @@ public class LevenbergMarquardtOptimizer implements LeastSquaresOptimizer {
final Incrementor iterationCounter = problem.getIterationCounter();
final Incrementor evaluationCounter = problem.getEvaluationCounter();
//convergence criterion
final ConvergenceChecker<PointVectorValuePair> checker
final ConvergenceChecker<Evaluation> checker
= problem.getConvergenceChecker();
// arrays shared with the other private methods
@ -319,7 +318,6 @@ public class LevenbergMarquardtOptimizer implements LeastSquaresOptimizer {
double[] diag = new double[nC];
double[] oldX = new double[nC];
double[] oldRes = new double[nR];
double[] oldObj = new double[nR];
double[] qtf = new double[nR];
double[] work1 = new double[nC];
double[] work2 = new double[nC];
@ -329,23 +327,20 @@ public class LevenbergMarquardtOptimizer implements LeastSquaresOptimizer {
// Evaluate the function at the starting point and calculate its norm.
evaluationCounter.incrementCount();
//value will be reassigned in the loop
Evaluation value = problem.evaluate(currentPoint);
double[] currentObjective = value.computeValue();
double[] currentResiduals = value.computeResiduals();
PointVectorValuePair current = new PointVectorValuePair(currentPoint, currentObjective);
double currentCost = value.computeCost();
Evaluation current = problem.evaluate(currentPoint);
double[] currentResiduals = current.computeResiduals();
double currentCost = current.computeCost();
// Outer loop.
boolean firstIteration = true;
while (true) {
iterationCounter.incrementCount();
final PointVectorValuePair previous = current;
final Evaluation previousValue = value;
final Evaluation previous = current;
// QR decomposition of the jacobian matrix
final InternalData internalData
= qrDecomposition(value.computeJacobian(), solvedCols);
= qrDecomposition(current.computeJacobian(), solvedCols);
final double[][] weightedJacobian = internalData.weightedJacobian;
final int[] permutation = internalData.permutation;
final double[] diagR = internalData.diagR;
@ -404,7 +399,7 @@ public class LevenbergMarquardtOptimizer implements LeastSquaresOptimizer {
if (maxCosine <= orthoTolerance) {
// Convergence has been reached.
return new OptimumImpl(
value,
current,
evaluationCounter.getCount(),
iterationCounter.getCount());
}
@ -426,9 +421,6 @@ public class LevenbergMarquardtOptimizer implements LeastSquaresOptimizer {
double[] tmpVec = weightedResidual;
weightedResidual = oldRes;
oldRes = tmpVec;
tmpVec = currentObjective;
currentObjective = oldObj;
oldObj = tmpVec;
// determine the Levenberg-Marquardt parameter
lmPar = determineLMParameter(qtf, delta, diag,
@ -452,11 +444,9 @@ public class LevenbergMarquardtOptimizer implements LeastSquaresOptimizer {
// Evaluate the function at x + p and calculate its norm.
evaluationCounter.incrementCount();
value = problem.evaluate(currentPoint);
currentObjective = value.computeValue();
currentResiduals = value.computeResiduals();
current = new PointVectorValuePair(currentPoint, currentObjective);
currentCost = value.computeCost();
current = problem.evaluate(currentPoint);
currentResiduals = current.computeResiduals();
currentCost = current.computeCost();
// compute the scaled actual reduction
double actRed = -1.0;
@ -515,7 +505,7 @@ public class LevenbergMarquardtOptimizer implements LeastSquaresOptimizer {
// tests for convergence.
if (checker != null && checker.converged(iterationCounter.getCount(), previous, current)) {
return new OptimumImpl(value, iterationCounter.getCount(), evaluationCounter.getCount());
return new OptimumImpl(current, iterationCounter.getCount(), evaluationCounter.getCount());
}
} else {
// failed iteration, reset the previous values
@ -527,12 +517,8 @@ public class LevenbergMarquardtOptimizer implements LeastSquaresOptimizer {
tmpVec = weightedResidual;
weightedResidual = oldRes;
oldRes = tmpVec;
tmpVec = currentObjective;
currentObjective = oldObj;
oldObj = tmpVec;
// Reset "current" to previous values.
current = new PointVectorValuePair(currentPoint, currentObjective);
value = previousValue;
current = previous;
}
// Default convergence criteria.
@ -540,7 +526,7 @@ public class LevenbergMarquardtOptimizer implements LeastSquaresOptimizer {
preRed <= costRelativeTolerance &&
ratio <= 2.0) ||
delta <= parRelativeTolerance * xNorm) {
return new OptimumImpl(value, iterationCounter.getCount(), evaluationCounter.getCount());
return new OptimumImpl(current, iterationCounter.getCount(), evaluationCounter.getCount());
}
// tests for termination and stringent tolerances

2
src/test/java/org/apache/commons/math3/fitting/leastsquares/AbstractLeastSquaresOptimizerAbstractTest.java
View File

@ -46,7 +46,7 @@ public abstract class AbstractLeastSquaresOptimizerAbstractTest {
public LeastSquaresBuilder base() {
return new LeastSquaresBuilder()
.checker(new SimpleVectorValueChecker(1e-6, 1e-6))
.checkerPair(new SimpleVectorValueChecker(1e-6, 1e-6))
.maxEvaluations(100)
.maxIterations(getMaxIterations());
}

2
src/test/java/org/apache/commons/math3/fitting/leastsquares/GaussNewtonOptimizerTest.java
View File

@ -96,7 +96,7 @@ public class GaussNewtonOptimizerTest
circle.addPoint( 45.0, 97.0);
LeastSquaresProblem lsp = builder(circle)
.checker(new SimpleVectorValueChecker(1e-30, 1e-30))
.checkerPair(new SimpleVectorValueChecker(1e-30, 1e-30))
.maxIterations(Integer.MAX_VALUE)
.start(new double[]{98.680, 47.345})
.build();