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

reverting commit introduced in r1426616 

git-svn-id: https://svn.apache.org/repos/asf/commons/proper/math/trunk@1426751 13f79535-47bb-0310-9956-ffa450edef68
This commit is contained in:
Luc Maisonobe 2012-12-29 12:10:52 +00:00
parent 250cf6e366
commit ed39d2dbe2
18 changed files with 145 additions and 297 deletions
Show Stats Download Patch File Download Diff File Expand all files Collapse all files

4
pom.xml
View File

@ -24,7 +24,7 @@
<modelVersion>4.0.0</modelVersion> <modelVersion>4.0.0</modelVersion>
<groupId>org.apache.commons</groupId> <groupId>org.apache.commons</groupId>
<artifactId>commons-math3</artifactId> <artifactId>commons-math3</artifactId>
<version>3.1.1-SNAPSHOT</version> <version>3.2-SNAPSHOT</version>
<name>Commons Math</name> <name>Commons Math</name>
<inceptionYear>2003</inceptionYear> <inceptionYear>2003</inceptionYear>
@ -293,7 +293,7 @@
<properties> <properties>
<commons.componentid>math3</commons.componentid> <commons.componentid>math3</commons.componentid>
<!-- do not use snapshot suffix here --> <!-- do not use snapshot suffix here -->
<commons.release.version>3.1.1</commons.release.version> <commons.release.version>3.2</commons.release.version>
<commons.release.desc>(requires Java 1.5+)</commons.release.desc> <commons.release.desc>(requires Java 1.5+)</commons.release.desc>
<!-- <commons.rc.version>RC1</commons.rc.version> --> <!-- <commons.rc.version>RC1</commons.rc.version> -->
<commons.binary.suffix>-bin</commons.binary.suffix> <commons.binary.suffix>-bin</commons.binary.suffix>

7
src/changes/changes.xml
View File

@ -50,13 +50,6 @@ If the output is not quite correct, check for invisible trailing spaces!
<title>Commons Math Release Notes</title> <title>Commons Math Release Notes</title>
</properties> </properties>
<body> <body>
<release version="3.1.1" date="TBD" description="
This is a micro release: It only contains bug fixes bug fixes.
">
<action dev="luc" type="fix" issue="MATH-924">
Fix handling of large number of weights in the new optimizers API.
</action>
</release>
<release version="3.1" date="2012-12-23" description=" <release version="3.1" date="2012-12-23" description="
This is a minor release: It combines bug fixes and new features. This is a minor release: It combines bug fixes and new features.
Changes to existing features were made in a backwards-compatible Changes to existing features were made in a backwards-compatible

11
src/main/java/org/apache/commons/math3/fitting/CurveFitter.java
View File

@ -18,18 +18,17 @@ package org.apache.commons.math3.fitting;
import java.util.ArrayList; import java.util.ArrayList;
import java.util.List; import java.util.List;
import org.apache.commons.math3.analysis.MultivariateMatrixFunction;
import org.apache.commons.math3.analysis.MultivariateVectorFunction; import org.apache.commons.math3.analysis.MultivariateVectorFunction;
import org.apache.commons.math3.analysis.MultivariateMatrixFunction;
import org.apache.commons.math3.analysis.ParametricUnivariateFunction; import org.apache.commons.math3.analysis.ParametricUnivariateFunction;
import org.apache.commons.math3.optim.InitialGuess;
import org.apache.commons.math3.optim.MaxEval; import org.apache.commons.math3.optim.MaxEval;
import org.apache.commons.math3.optim.InitialGuess;
import org.apache.commons.math3.optim.PointVectorValuePair; import org.apache.commons.math3.optim.PointVectorValuePair;
import org.apache.commons.math3.optim.nonlinear.vector.MultivariateVectorOptimizer;
import org.apache.commons.math3.optim.nonlinear.vector.ModelFunction; import org.apache.commons.math3.optim.nonlinear.vector.ModelFunction;
import org.apache.commons.math3.optim.nonlinear.vector.ModelFunctionJacobian; import org.apache.commons.math3.optim.nonlinear.vector.ModelFunctionJacobian;
import org.apache.commons.math3.optim.nonlinear.vector.MultivariateVectorOptimizer;
import org.apache.commons.math3.optim.nonlinear.vector.Target; import org.apache.commons.math3.optim.nonlinear.vector.Target;
import org.apache.commons.math3.optim.nonlinear.vector.NonCorrelatedWeight; import org.apache.commons.math3.optim.nonlinear.vector.Weight;
/** /**
* Fitter for parametric univariate real functions y = f(x). * Fitter for parametric univariate real functions y = f(x).
@ -175,7 +174,7 @@ public class CurveFitter<T extends ParametricUnivariateFunction> {
model.getModelFunction(), model.getModelFunction(),
model.getModelFunctionJacobian(), model.getModelFunctionJacobian(),
new Target(target), new Target(target),
new NonCorrelatedWeight(weights), new Weight(weights),
new InitialGuess(initialGuess)); new InitialGuess(initialGuess));
// Extract the coefficients. // Extract the coefficients.
return optimum.getPointRef(); return optimum.getPointRef();

19
src/main/java/org/apache/commons/math3/optim/nonlinear/vector/MultiStartMultivariateVectorOptimizer.java
View File

@ -16,18 +16,18 @@
*/ */
package org.apache.commons.math3.optim.nonlinear.vector; package org.apache.commons.math3.optim.nonlinear.vector;
import java.util.ArrayList;
import java.util.Collections; import java.util.Collections;
import java.util.Comparator;
import java.util.List; import java.util.List;
import java.util.ArrayList;
import java.util.Comparator;
import org.apache.commons.math3.exception.NotStrictlyPositiveException; import org.apache.commons.math3.exception.NotStrictlyPositiveException;
import org.apache.commons.math3.exception.NullArgumentException; import org.apache.commons.math3.exception.NullArgumentException;
import org.apache.commons.math3.linear.ArrayRealVector; import org.apache.commons.math3.linear.RealMatrix;
import org.apache.commons.math3.linear.RealVector; import org.apache.commons.math3.linear.RealVector;
import org.apache.commons.math3.linear.ArrayRealVector;
import org.apache.commons.math3.random.RandomVectorGenerator;
import org.apache.commons.math3.optim.BaseMultiStartMultivariateOptimizer; import org.apache.commons.math3.optim.BaseMultiStartMultivariateOptimizer;
import org.apache.commons.math3.optim.PointVectorValuePair; import org.apache.commons.math3.optim.PointVectorValuePair;
import org.apache.commons.math3.random.RandomVectorGenerator;
/** /**
* Multi-start optimizer for a (vector) model function. * Multi-start optimizer for a (vector) model function.
@ -98,7 +98,7 @@ public class MultiStartMultivariateVectorOptimizer
private Comparator<PointVectorValuePair> getPairComparator() { private Comparator<PointVectorValuePair> getPairComparator() {
return new Comparator<PointVectorValuePair>() { return new Comparator<PointVectorValuePair>() {
private final RealVector target = new ArrayRealVector(optimizer.getTarget(), false); private final RealVector target = new ArrayRealVector(optimizer.getTarget(), false);
private final double[] weight = optimizer.getNonCorrelatedWeight(); private final RealMatrix weight = optimizer.getWeight();
public int compare(final PointVectorValuePair o1, public int compare(final PointVectorValuePair o1,
final PointVectorValuePair o2) { final PointVectorValuePair o2) {
@ -114,12 +114,7 @@ public class MultiStartMultivariateVectorOptimizer
private double weightedResidual(final PointVectorValuePair pv) { private double weightedResidual(final PointVectorValuePair pv) {
final RealVector v = new ArrayRealVector(pv.getValueRef(), false); final RealVector v = new ArrayRealVector(pv.getValueRef(), false);
final RealVector r = target.subtract(v); final RealVector r = target.subtract(v);
double sum = 0; return r.dotProduct(weight.operate(r));
for (int i = 0; i < r.getDimension(); ++i) {
final double ri = r.getEntry(i);
sum += ri * weight[i] * ri;
}
return sum;
} }
}; };
} }

62
src/main/java/org/apache/commons/math3/optim/nonlinear/vector/MultivariateVectorOptimizer.java
View File

@ -17,15 +17,14 @@
package org.apache.commons.math3.optim.nonlinear.vector; package org.apache.commons.math3.optim.nonlinear.vector;
import org.apache.commons.math3.analysis.MultivariateVectorFunction;
import org.apache.commons.math3.exception.DimensionMismatchException;
import org.apache.commons.math3.exception.TooManyEvaluationsException; import org.apache.commons.math3.exception.TooManyEvaluationsException;
import org.apache.commons.math3.linear.RealMatrix; import org.apache.commons.math3.exception.DimensionMismatchException;
import org.apache.commons.math3.analysis.MultivariateVectorFunction;
import org.apache.commons.math3.optim.OptimizationData;
import org.apache.commons.math3.optim.BaseMultivariateOptimizer; import org.apache.commons.math3.optim.BaseMultivariateOptimizer;
import org.apache.commons.math3.optim.ConvergenceChecker; import org.apache.commons.math3.optim.ConvergenceChecker;
import org.apache.commons.math3.optim.OptimizationData;
import org.apache.commons.math3.optim.PointVectorValuePair; import org.apache.commons.math3.optim.PointVectorValuePair;
import org.apache.commons.math3.optim.nonlinear.vector.jacobian.GaussNewtonOptimizer; import org.apache.commons.math3.linear.RealMatrix;
/** /**
* Base class for a multivariate vector function optimizer. * Base class for a multivariate vector function optimizer.
@ -37,13 +36,8 @@ public abstract class MultivariateVectorOptimizer
extends BaseMultivariateOptimizer<PointVectorValuePair> { extends BaseMultivariateOptimizer<PointVectorValuePair> {
/** Target values for the model function at optimum. */ /** Target values for the model function at optimum. */
private double[] target; private double[] target;
/** Weight matrix. /** Weight matrix. */
* @deprecated as of 3.1.1, replaced by weight
*/
@Deprecated
private RealMatrix weightMatrix; private RealMatrix weightMatrix;
/** Weight vector. */
private double[] weight;
/** Model function. */ /** Model function. */
private MultivariateVectorFunction model; private MultivariateVectorFunction model;
@ -71,25 +65,14 @@ public abstract class MultivariateVectorOptimizer
/** /**
* {@inheritDoc} * {@inheritDoc}
* <p> *
* Note that for version 3.1 of Apache Commons Math, a general <code>Weight</code>
* data was looked for, which could hold arbitrary square matrices and not only
* vector as the current {@link NonCorrelatedWeight} does. This was flawed as some
* optimizers like {@link GaussNewtonOptimizer} only considered the diagonal elements.
* This feature was deprecated. If users need non-diagonal weights to handle correlated
* observations, they will have to implement it by themselves using pre-multiplication
* by a matrix in both their function implementation and observation vectors. There is
* no direct support for this anymore in the Apache Commons Math library. The only
* feature that is supported here is a convenience feature for non-correlated observations,
* with vector only weights (i.e. weight[i] is the weight for observation i).
* </p>
* @param optData Optimization data. The following data will be looked for: * @param optData Optimization data. The following data will be looked for:
* <ul> * <ul>
* <li>{@link org.apache.commons.math3.optim.MaxEval}</li> * <li>{@link org.apache.commons.math3.optim.MaxEval}</li>
* <li>{@link org.apache.commons.math3.optim.InitialGuess}</li> * <li>{@link org.apache.commons.math3.optim.InitialGuess}</li>
* <li>{@link org.apache.commons.math3.optim.SimpleBounds}</li> * <li>{@link org.apache.commons.math3.optim.SimpleBounds}</li>
* <li>{@link Target}</li> * <li>{@link Target}</li>
* <li>{@link NonCorrelatedWeight}</li> * <li>{@link Weight}</li>
* <li>{@link ModelFunction}</li> * <li>{@link ModelFunction}</li>
* </ul> * </ul>
* @return {@inheritDoc} * @return {@inheritDoc}
@ -113,22 +96,10 @@ public abstract class MultivariateVectorOptimizer
* Gets the weight matrix of the observations. * Gets the weight matrix of the observations.
* *
* @return the weight matrix. * @return the weight matrix.
* @deprecated as of 3.1.1, replaced by {@link #getNonCorrelatedWeight()}
*/ */
@Deprecated
public RealMatrix getWeight() { public RealMatrix getWeight() {
return weightMatrix.copy(); return weightMatrix.copy();
} }
/**
* Gets the weights of the observations.
*
* @return the weights.
* @since 3.1.1
*/
public double[] getNonCorrelatedWeight() {
return weight.clone();
}
/** /**
* Gets the observed values to be matched by the objective vector * Gets the observed values to be matched by the objective vector
* function. * function.
@ -155,7 +126,7 @@ public abstract class MultivariateVectorOptimizer
* @param optData Optimization data. The following data will be looked for: * @param optData Optimization data. The following data will be looked for:
* <ul> * <ul>
* <li>{@link Target}</li> * <li>{@link Target}</li>
* <li>{@link NonCorrelatedWeight}</li> * <li>{@link Weight}</li>
* <li>{@link ModelFunction}</li> * <li>{@link ModelFunction}</li>
* </ul> * </ul>
*/ */
@ -171,18 +142,8 @@ public abstract class MultivariateVectorOptimizer
target = ((Target) data).getTarget(); target = ((Target) data).getTarget();
continue; continue;
} }
if (data instanceof NonCorrelatedWeight) {
weight = ((NonCorrelatedWeight) data).getWeight();
continue;
}
// TODO: remove this for 4.0, when the Weight class will be removed
if (data instanceof Weight) { if (data instanceof Weight) {
weightMatrix = ((Weight) data).getWeight(); weightMatrix = ((Weight) data).getWeight();
weight = new double[weightMatrix.getColumnDimension()];
for (int i = 0; i < weight.length; ++i) {
// extract the diagonal of the matrix
weight[i] = weightMatrix.getEntry(i, i);
}
continue; continue;
} }
} }
@ -192,11 +153,12 @@ public abstract class MultivariateVectorOptimizer
* Check parameters consistency. * Check parameters consistency.
* *
* @throws DimensionMismatchException if {@link #target} and * @throws DimensionMismatchException if {@link #target} and
* {@link #weight} have inconsistent dimensions. * {@link #weightMatrix} have inconsistent dimensions.
*/ */
private void checkParameters() { private void checkParameters() {
if (target.length != weight.length) { if (target.length != weightMatrix.getColumnDimension()) {
throw new DimensionMismatchException(target.length, weight.length); throw new DimensionMismatchException(target.length,
weightMatrix.getColumnDimension());
} }
} }
} }

53
src/main/java/org/apache/commons/math3/optim/nonlinear/vector/NonCorrelatedWeight.java
View File

@ -1,53 +0,0 @@
/*
* Licensed to the Apache Software Foundation (ASF) under one or more
* contributor license agreements. See the NOTICE file distributed with
* this work for additional information regarding copyright ownership.
* The ASF licenses this file to You 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 org.apache.commons.math3.optim.nonlinear.vector;
import org.apache.commons.math3.optim.OptimizationData;
/**
* Weight of the residuals between model and observations, when
* observations are non-correlated.
* <br/>
* Immutable class.
*
* @version $Id$
* @since 3.1.1
*/
public class NonCorrelatedWeight implements OptimizationData {
/** Weight. */
private final double[] weight;
/**
* Creates a weight vector.
*
* @param weight weight of the observations
*/
public NonCorrelatedWeight(final double[] weight) {
this.weight = weight.clone();
}
/**
* Gets the weight.
*
* @return a fresh copy of the weight.
*/
public double[] getWeight() {
return weight.clone();
}
}

18
src/main/java/org/apache/commons/math3/optim/nonlinear/vector/Weight.java
View File

@ -28,20 +28,22 @@ import org.apache.commons.math3.linear.NonSquareMatrixException;
* *
* @version $Id: Weight.java 1416643 2012-12-03 19:37:14Z tn $ * @version $Id: Weight.java 1416643 2012-12-03 19:37:14Z tn $
* @since 3.1 * @since 3.1
* @deprecated as of 3.1.1, replaced by {@link NonCorrelatedWeight}
*/ */
@Deprecated
public class Weight implements OptimizationData { public class Weight implements OptimizationData {
/** Weight matrix. */ /** Weight matrix. */
private final RealMatrix weightMatrix; private final RealMatrix weightMatrix;
/** /**
* Creates a weight matrix. * Creates a diagonal weight matrix.
* *
* @param weight matrix elements. * @param weight List of the values of the diagonal.
*/ */
public Weight(double[][] weight) { public Weight(double[] weight) {
weightMatrix = MatrixUtils.createRealMatrix(weight); final int dim = weight.length;
weightMatrix = MatrixUtils.createRealMatrix(dim, dim);
for (int i = 0; i < dim; i++) {
weightMatrix.setEntry(i, i, weight[i]);
}
} }
/** /**
@ -59,9 +61,9 @@ public class Weight implements OptimizationData {
} }
/** /**
* Gets the weight. * Gets the initial guess.
* *
* @return a fresh copy of the weight. * @return the initial guess.
*/ */
public RealMatrix getWeight() { public RealMatrix getWeight() {
return weightMatrix.copy(); return weightMatrix.copy();

75
src/main/java/org/apache/commons/math3/optim/nonlinear/vector/jacobian/AbstractLeastSquaresOptimizer.java
View File

@ -19,18 +19,16 @@ package org.apache.commons.math3.optim.nonlinear.vector.jacobian;
import org.apache.commons.math3.exception.DimensionMismatchException; import org.apache.commons.math3.exception.DimensionMismatchException;
import org.apache.commons.math3.exception.TooManyEvaluationsException; import org.apache.commons.math3.exception.TooManyEvaluationsException;
import org.apache.commons.math3.linear.ArrayRealVector; import org.apache.commons.math3.linear.ArrayRealVector;
import org.apache.commons.math3.linear.RealMatrix;
import org.apache.commons.math3.linear.DecompositionSolver; import org.apache.commons.math3.linear.DecompositionSolver;
import org.apache.commons.math3.linear.EigenDecomposition;
import org.apache.commons.math3.linear.MatrixUtils; import org.apache.commons.math3.linear.MatrixUtils;
import org.apache.commons.math3.linear.QRDecomposition; import org.apache.commons.math3.linear.QRDecomposition;
import org.apache.commons.math3.linear.RealMatrix; import org.apache.commons.math3.linear.EigenDecomposition;
import org.apache.commons.math3.optim.ConvergenceChecker;
import org.apache.commons.math3.optim.OptimizationData; import org.apache.commons.math3.optim.OptimizationData;
import org.apache.commons.math3.optim.ConvergenceChecker;
import org.apache.commons.math3.optim.PointVectorValuePair; import org.apache.commons.math3.optim.PointVectorValuePair;
import org.apache.commons.math3.optim.nonlinear.vector.JacobianMultivariateVectorOptimizer;
import org.apache.commons.math3.optim.nonlinear.vector.MultivariateVectorOptimizer;
import org.apache.commons.math3.optim.nonlinear.vector.NonCorrelatedWeight;
import org.apache.commons.math3.optim.nonlinear.vector.Weight; import org.apache.commons.math3.optim.nonlinear.vector.Weight;
import org.apache.commons.math3.optim.nonlinear.vector.JacobianMultivariateVectorOptimizer;
import org.apache.commons.math3.util.FastMath; import org.apache.commons.math3.util.FastMath;
/** /**
@ -42,13 +40,8 @@ import org.apache.commons.math3.util.FastMath;
*/ */
public abstract class AbstractLeastSquaresOptimizer public abstract class AbstractLeastSquaresOptimizer
extends JacobianMultivariateVectorOptimizer { extends JacobianMultivariateVectorOptimizer {
/** Square-root of the weight matrix. /** Square-root of the weight matrix. */
* @deprecated as of 3.1.1, replaced by {@link #weight}
*/
@Deprecated
private RealMatrix weightMatrixSqrt; private RealMatrix weightMatrixSqrt;
/** Square-root of the weight vector. */
private double[] weightSquareRoot;
/** Cost value (square root of the sum of the residuals). */ /** Cost value (square root of the sum of the residuals). */
private double cost; private double cost;
@ -68,23 +61,7 @@ public abstract class AbstractLeastSquaresOptimizer
* match problem dimension. * match problem dimension.
*/ */
protected RealMatrix computeWeightedJacobian(double[] params) { protected RealMatrix computeWeightedJacobian(double[] params) {
return weightMatrixSqrt.multiply(MatrixUtils.createRealMatrix(computeJacobian(params)));
final double[][] jacobian = computeJacobian(params);
if (weightSquareRoot != null) {
for (int i = 0; i < jacobian.length; ++i) {
final double wi = weightSquareRoot[i];
final double[] row = jacobian[i];
for (int j = 0; j < row.length; ++j) {
row[j] *= wi;
}
}
return MatrixUtils.createRealMatrix(jacobian);
} else {
// TODO: remove for 4.0, when the {@link Weight} class will be removed
return weightMatrixSqrt.multiply(MatrixUtils.createRealMatrix(jacobian));
}
} }
/** /**
@ -96,13 +73,7 @@ public abstract class AbstractLeastSquaresOptimizer
*/ */
protected double computeCost(double[] residuals) { protected double computeCost(double[] residuals) {
final ArrayRealVector r = new ArrayRealVector(residuals); final ArrayRealVector r = new ArrayRealVector(residuals);
final double[] weight = getNonCorrelatedWeight(); return FastMath.sqrt(r.dotProduct(getWeight().operate(r)));
double sum = 0;
for (int i = 0; i < r.getDimension(); ++i) {
final double ri = r.getEntry(i);
sum += ri * weight[i] * ri;
}
return FastMath.sqrt(sum);
} }
/** /**
@ -134,9 +105,7 @@ public abstract class AbstractLeastSquaresOptimizer
* Gets the square-root of the weight matrix. * Gets the square-root of the weight matrix.
* *
* @return the square-root of the weight matrix. * @return the square-root of the weight matrix.
* @deprecated as of 3.1.1, replaced with {@link MultivariateVectorOptimizer#getNonCorrelatedWeight()}
*/ */
@Deprecated
public RealMatrix getWeightSquareRoot() { public RealMatrix getWeightSquareRoot() {
return weightMatrixSqrt.copy(); return weightMatrixSqrt.copy();
} }
@ -214,7 +183,7 @@ public abstract class AbstractLeastSquaresOptimizer
* <li>{@link org.apache.commons.math3.optim.InitialGuess}</li> * <li>{@link org.apache.commons.math3.optim.InitialGuess}</li>
* <li>{@link org.apache.commons.math3.optim.SimpleBounds}</li> * <li>{@link org.apache.commons.math3.optim.SimpleBounds}</li>
* <li>{@link org.apache.commons.math3.optim.nonlinear.vector.Target}</li> * <li>{@link org.apache.commons.math3.optim.nonlinear.vector.Target}</li>
* <li>{@link org.apache.commons.math3.optim.nonlinear.vector.NonCorrelatedWeight}</li> * <li>{@link org.apache.commons.math3.optim.nonlinear.vector.Weight}</li>
* <li>{@link org.apache.commons.math3.optim.nonlinear.vector.ModelFunction}</li> * <li>{@link org.apache.commons.math3.optim.nonlinear.vector.ModelFunction}</li>
* <li>{@link org.apache.commons.math3.optim.nonlinear.vector.ModelFunctionJacobian}</li> * <li>{@link org.apache.commons.math3.optim.nonlinear.vector.ModelFunctionJacobian}</li>
* </ul> * </ul>
@ -266,7 +235,8 @@ public abstract class AbstractLeastSquaresOptimizer
/** /**
* Scans the list of (required and optional) optimization data that * Scans the list of (required and optional) optimization data that
* characterize the problem. * characterize the problem.
* If the weight is specified, the {@link #weightSquareRoot} field is recomputed. * If the weight matrix is specified, the {@link #weightMatrixSqrt}
* field is recomputed.
* *
* @param optData Optimization data. The following data will be looked for: * @param optData Optimization data. The following data will be looked for:
* <ul> * <ul>
@ -278,19 +248,22 @@ public abstract class AbstractLeastSquaresOptimizer
// not provided in the argument list. // not provided in the argument list.
for (OptimizationData data : optData) { for (OptimizationData data : optData) {
if (data instanceof Weight) { if (data instanceof Weight) {
// TODO: remove for 4.0, when the {@link Weight} class will be removed weightMatrixSqrt = squareRoot(((Weight) data).getWeight());
weightSquareRoot = null; // If more data must be parsed, this statement _must_ be
final RealMatrix w = ((Weight) data).getWeight(); // changed to "continue".
final EigenDecomposition dec = new EigenDecomposition(w); break;
weightMatrixSqrt = dec.getSquareRoot();
} else if (data instanceof NonCorrelatedWeight) {
weightSquareRoot = ((NonCorrelatedWeight) data).getWeight();
for (int i = 0; i < weightSquareRoot.length; ++i) {
weightSquareRoot[i] = FastMath.sqrt(weightSquareRoot[i]);
}
weightMatrixSqrt = null;
} }
} }
} }
/**
* Computes the square-root of the weight matrix.
*
* @param m Symmetric, positive-definite (weight) matrix.
* @return the square-root of the weight matrix.
*/
private RealMatrix squareRoot(RealMatrix m) {
final EigenDecomposition dec = new EigenDecomposition(m);
return dec.getSquareRoot();
}
} }

9
src/main/java/org/apache/commons/math3/optim/nonlinear/vector/jacobian/GaussNewtonOptimizer.java
View File

@ -17,8 +17,8 @@
package org.apache.commons.math3.optim.nonlinear.vector.jacobian; package org.apache.commons.math3.optim.nonlinear.vector.jacobian;
import org.apache.commons.math3.exception.ConvergenceException; import org.apache.commons.math3.exception.ConvergenceException;
import org.apache.commons.math3.exception.MathInternalError;
import org.apache.commons.math3.exception.NullArgumentException; import org.apache.commons.math3.exception.NullArgumentException;
import org.apache.commons.math3.exception.MathInternalError;
import org.apache.commons.math3.exception.util.LocalizedFormats; import org.apache.commons.math3.exception.util.LocalizedFormats;
import org.apache.commons.math3.linear.ArrayRealVector; import org.apache.commons.math3.linear.ArrayRealVector;
import org.apache.commons.math3.linear.BlockRealMatrix; import org.apache.commons.math3.linear.BlockRealMatrix;
@ -83,7 +83,12 @@ public class GaussNewtonOptimizer extends AbstractLeastSquaresOptimizer {
final double[] targetValues = getTarget(); final double[] targetValues = getTarget();
final int nR = targetValues.length; // Number of observed data. final int nR = targetValues.length; // Number of observed data.
final double[] residualsWeights = getNonCorrelatedWeight(); final RealMatrix weightMatrix = getWeight();
// Diagonal of the weight matrix.
final double[] residualsWeights = new double[nR];
for (int i = 0; i < nR; i++) {
residualsWeights[i] = weightMatrix.getEntry(i, i);
}
final double[] currentPoint = getStartPoint(); final double[] currentPoint = getStartPoint();
final int nC = currentPoint.length; final int nC = currentPoint.length;

14
src/main/java/org/apache/commons/math3/optim/nonlinear/vector/jacobian/LevenbergMarquardtOptimizer.java
View File

@ -17,14 +17,13 @@
package org.apache.commons.math3.optim.nonlinear.vector.jacobian; package org.apache.commons.math3.optim.nonlinear.vector.jacobian;
import java.util.Arrays; import java.util.Arrays;
import org.apache.commons.math3.exception.ConvergenceException; import org.apache.commons.math3.exception.ConvergenceException;
import org.apache.commons.math3.exception.util.LocalizedFormats; import org.apache.commons.math3.exception.util.LocalizedFormats;
import org.apache.commons.math3.linear.RealMatrix;
import org.apache.commons.math3.optim.ConvergenceChecker;
import org.apache.commons.math3.optim.PointVectorValuePair; import org.apache.commons.math3.optim.PointVectorValuePair;
import org.apache.commons.math3.util.FastMath; import org.apache.commons.math3.optim.ConvergenceChecker;
import org.apache.commons.math3.linear.RealMatrix;
import org.apache.commons.math3.util.Precision; import org.apache.commons.math3.util.Precision;
import org.apache.commons.math3.util.FastMath;
/** /**
@ -301,7 +300,7 @@ public class LevenbergMarquardtOptimizer
double[] work2 = new double[nC]; double[] work2 = new double[nC];
double[] work3 = new double[nC]; double[] work3 = new double[nC];
final double[] weight = getNonCorrelatedWeight(); final RealMatrix weightMatrixSqrt = getWeightSquareRoot();
// Evaluate the function at the starting point and calculate its norm. // Evaluate the function at the starting point and calculate its norm.
double[] currentObjective = computeObjectiveValue(currentPoint); double[] currentObjective = computeObjectiveValue(currentPoint);
@ -321,10 +320,7 @@ public class LevenbergMarquardtOptimizer
// QR decomposition of the jacobian matrix // QR decomposition of the jacobian matrix
qrDecomposition(computeWeightedJacobian(currentPoint)); qrDecomposition(computeWeightedJacobian(currentPoint));
weightedResidual = new double[currentResiduals.length]; weightedResidual = weightMatrixSqrt.operate(currentResiduals);
for (int i = 0; i < weightedResidual.length; ++i) {
weightedResidual[i] = FastMath.sqrt(weight[i]) * currentResiduals[i];
}
for (int i = 0; i < nR; i++) { for (int i = 0; i < nR; i++) {
qtf[i] = weightedResidual[i]; qtf[i] = weightedResidual[i];
} }

27
src/test/java/org/apache/commons/math3/fitting/PolynomialFitterTest.java
View File

@ -220,33 +220,6 @@ public class PolynomialFitterTest {
checkUnsolvableProblem(new GaussNewtonOptimizer(true, new SimpleVectorValueChecker(1e-15, 1e-15)), false); checkUnsolvableProblem(new GaussNewtonOptimizer(true, new SimpleVectorValueChecker(1e-15, 1e-15)), false);
} }
@Test
public void testLargeSample() {
Random randomizer = new Random(0x5551480dca5b369bl);
double maxError = 0;
for (int degree = 0; degree < 10; ++degree) {
PolynomialFunction p = buildRandomPolynomial(degree, randomizer);
PolynomialFitter fitter = new PolynomialFitter(new LevenbergMarquardtOptimizer());
for (int i = 0; i < 40000; ++i) {
double x = -1.0 + i / 20000.0;
fitter.addObservedPoint(1.0, x,
p.value(x) + 0.1 * randomizer.nextGaussian());
}
final double[] init = new double[degree + 1];
PolynomialFunction fitted = new PolynomialFunction(fitter.fit(init));
for (double x = -1.0; x < 1.0; x += 0.01) {
double error = FastMath.abs(p.value(x) - fitted.value(x)) /
(1.0 + FastMath.abs(p.value(x)));
maxError = FastMath.max(maxError, error);
Assert.assertTrue(FastMath.abs(error) < 0.01);
}
}
Assert.assertTrue(maxError > 0.001);
}
private void checkUnsolvableProblem(MultivariateVectorOptimizer optimizer, private void checkUnsolvableProblem(MultivariateVectorOptimizer optimizer,
boolean solvable) { boolean solvable) {
Random randomizer = new Random(1248788532l); Random randomizer = new Random(1248788532l);

9
src/test/java/org/apache/commons/math3/optim/nonlinear/vector/MultiStartMultivariateVectorOptimizerTest.java
View File

@ -16,12 +16,13 @@
*/ */
package org.apache.commons.math3.optim.nonlinear.vector; package org.apache.commons.math3.optim.nonlinear.vector;
import org.apache.commons.math3.analysis.MultivariateMatrixFunction;
import org.apache.commons.math3.analysis.MultivariateVectorFunction; import org.apache.commons.math3.analysis.MultivariateVectorFunction;
import org.apache.commons.math3.analysis.MultivariateMatrixFunction;
import org.apache.commons.math3.exception.MathIllegalStateException;
import org.apache.commons.math3.linear.BlockRealMatrix; import org.apache.commons.math3.linear.BlockRealMatrix;
import org.apache.commons.math3.linear.RealMatrix; import org.apache.commons.math3.linear.RealMatrix;
import org.apache.commons.math3.optim.InitialGuess;
import org.apache.commons.math3.optim.MaxEval; import org.apache.commons.math3.optim.MaxEval;
import org.apache.commons.math3.optim.InitialGuess;
import org.apache.commons.math3.optim.PointVectorValuePair; import org.apache.commons.math3.optim.PointVectorValuePair;
import org.apache.commons.math3.optim.SimpleVectorValueChecker; import org.apache.commons.math3.optim.SimpleVectorValueChecker;
import org.apache.commons.math3.optim.nonlinear.vector.jacobian.GaussNewtonOptimizer; import org.apache.commons.math3.optim.nonlinear.vector.jacobian.GaussNewtonOptimizer;
@ -129,7 +130,7 @@ public class MultiStartMultivariateVectorOptimizerTest {
problem.getModelFunction(), problem.getModelFunction(),
problem.getModelFunctionJacobian(), problem.getModelFunctionJacobian(),
problem.getTarget(), problem.getTarget(),
new NonCorrelatedWeight(new double[] { 1 }), new Weight(new double[] { 1 }),
new InitialGuess(new double[] { 0 })); new InitialGuess(new double[] { 0 }));
Assert.assertEquals(1.5, optimum.getPoint()[0], 1e-10); Assert.assertEquals(1.5, optimum.getPoint()[0], 1e-10);
Assert.assertEquals(3.0, optimum.getValue()[0], 1e-10); Assert.assertEquals(3.0, optimum.getValue()[0], 1e-10);
@ -160,7 +161,7 @@ public class MultiStartMultivariateVectorOptimizerTest {
= new MultiStartMultivariateVectorOptimizer(underlyingOptimizer, 10, generator); = new MultiStartMultivariateVectorOptimizer(underlyingOptimizer, 10, generator);
optimizer.optimize(new MaxEval(100), optimizer.optimize(new MaxEval(100),
new Target(new double[] { 0 }), new Target(new double[] { 0 }),
new NonCorrelatedWeight(new double[] { 1 }), new Weight(new double[] { 1 }),
new InitialGuess(new double[] { 0 }), new InitialGuess(new double[] { 0 }),
new ModelFunction(new MultivariateVectorFunction() { new ModelFunction(new MultivariateVectorFunction() {
public double[] value(double[] point) { public double[] value(double[] point) {

53
src/test/java/org/apache/commons/math3/optim/nonlinear/vector/jacobian/AbstractLeastSquaresOptimizerAbstractTest.java
View File

@ -17,22 +17,23 @@
package org.apache.commons.math3.optim.nonlinear.vector.jacobian; package org.apache.commons.math3.optim.nonlinear.vector.jacobian;
import java.io.IOException; import java.io.IOException;
import java.io.Serializable;
import java.util.Arrays; import java.util.Arrays;
import org.apache.commons.math3.analysis.MultivariateMatrixFunction;
import org.apache.commons.math3.analysis.MultivariateVectorFunction; import org.apache.commons.math3.analysis.MultivariateVectorFunction;
import org.apache.commons.math3.analysis.MultivariateMatrixFunction;
import org.apache.commons.math3.exception.ConvergenceException; import org.apache.commons.math3.exception.ConvergenceException;
import org.apache.commons.math3.exception.DimensionMismatchException; import org.apache.commons.math3.exception.DimensionMismatchException;
import org.apache.commons.math3.exception.NumberIsTooSmallException;
import org.apache.commons.math3.geometry.euclidean.twod.Vector2D; import org.apache.commons.math3.geometry.euclidean.twod.Vector2D;
import org.apache.commons.math3.linear.BlockRealMatrix; import org.apache.commons.math3.linear.BlockRealMatrix;
import org.apache.commons.math3.linear.RealMatrix; import org.apache.commons.math3.linear.RealMatrix;
import org.apache.commons.math3.optim.PointVectorValuePair;
import org.apache.commons.math3.optim.InitialGuess; import org.apache.commons.math3.optim.InitialGuess;
import org.apache.commons.math3.optim.MaxEval; import org.apache.commons.math3.optim.MaxEval;
import org.apache.commons.math3.optim.PointVectorValuePair; import org.apache.commons.math3.optim.nonlinear.vector.Target;
import org.apache.commons.math3.optim.nonlinear.vector.Weight;
import org.apache.commons.math3.optim.nonlinear.vector.ModelFunction; import org.apache.commons.math3.optim.nonlinear.vector.ModelFunction;
import org.apache.commons.math3.optim.nonlinear.vector.ModelFunctionJacobian; import org.apache.commons.math3.optim.nonlinear.vector.ModelFunctionJacobian;
import org.apache.commons.math3.optim.nonlinear.vector.Target;
import org.apache.commons.math3.optim.nonlinear.vector.NonCorrelatedWeight;
import org.apache.commons.math3.util.FastMath; import org.apache.commons.math3.util.FastMath;
import org.junit.Assert; import org.junit.Assert;
import org.junit.Test; import org.junit.Test;
@ -114,7 +115,7 @@ public abstract class AbstractLeastSquaresOptimizerAbstractTest {
problem.getModelFunction(), problem.getModelFunction(),
problem.getModelFunctionJacobian(), problem.getModelFunctionJacobian(),
problem.getTarget(), problem.getTarget(),
new NonCorrelatedWeight(new double[] { 1 }), new Weight(new double[] { 1 }),
new InitialGuess(new double[] { 0 })); new InitialGuess(new double[] { 0 }));
Assert.assertEquals(0, optimizer.getRMS(), 1e-10); Assert.assertEquals(0, optimizer.getRMS(), 1e-10);
Assert.assertEquals(1.5, optimum.getPoint()[0], 1e-10); Assert.assertEquals(1.5, optimum.getPoint()[0], 1e-10);
@ -134,7 +135,7 @@ public abstract class AbstractLeastSquaresOptimizerAbstractTest {
problem.getModelFunction(), problem.getModelFunction(),
problem.getModelFunctionJacobian(), problem.getModelFunctionJacobian(),
problem.getTarget(), problem.getTarget(),
new NonCorrelatedWeight(new double[] { 1, 1, 1 }), new Weight(new double[] { 1, 1, 1 }),
new InitialGuess(new double[] { 0, 0 })); new InitialGuess(new double[] { 0, 0 }));
Assert.assertEquals(0, optimizer.getRMS(), 1e-10); Assert.assertEquals(0, optimizer.getRMS(), 1e-10);
Assert.assertEquals(7, optimum.getPoint()[0], 1e-10); Assert.assertEquals(7, optimum.getPoint()[0], 1e-10);
@ -160,7 +161,7 @@ public abstract class AbstractLeastSquaresOptimizerAbstractTest {
problem.getModelFunction(), problem.getModelFunction(),
problem.getModelFunctionJacobian(), problem.getModelFunctionJacobian(),
problem.getTarget(), problem.getTarget(),
new NonCorrelatedWeight(new double[] { 1, 1, 1, 1, 1, 1 }), new Weight(new double[] { 1, 1, 1, 1, 1, 1 }),
new InitialGuess(new double[] { 0, 0, 0, 0, 0, 0 })); new InitialGuess(new double[] { 0, 0, 0, 0, 0, 0 }));
Assert.assertEquals(0, optimizer.getRMS(), 1e-10); Assert.assertEquals(0, optimizer.getRMS(), 1e-10);
for (int i = 0; i < problem.target.length; ++i) { for (int i = 0; i < problem.target.length; ++i) {
@ -182,7 +183,7 @@ public abstract class AbstractLeastSquaresOptimizerAbstractTest {
problem.getModelFunction(), problem.getModelFunction(),
problem.getModelFunctionJacobian(), problem.getModelFunctionJacobian(),
problem.getTarget(), problem.getTarget(),
new NonCorrelatedWeight(new double[] { 1, 1, 1 }), new Weight(new double[] { 1, 1, 1 }),
new InitialGuess(new double[] { 0, 0, 0 })); new InitialGuess(new double[] { 0, 0, 0 }));
Assert.assertEquals(0, optimizer.getRMS(), 1e-10); Assert.assertEquals(0, optimizer.getRMS(), 1e-10);
Assert.assertEquals(1, optimum.getPoint()[0], 1e-10); Assert.assertEquals(1, optimum.getPoint()[0], 1e-10);
@ -208,7 +209,7 @@ public abstract class AbstractLeastSquaresOptimizerAbstractTest {
problem.getModelFunction(), problem.getModelFunction(),
problem.getModelFunctionJacobian(), problem.getModelFunctionJacobian(),
problem.getTarget(), problem.getTarget(),
new NonCorrelatedWeight(new double[] { 1, 1, 1, 1, 1, 1 }), new Weight(new double[] { 1, 1, 1, 1, 1, 1 }),
new InitialGuess(new double[] { 0, 0, 0, 0, 0, 0 })); new InitialGuess(new double[] { 0, 0, 0, 0, 0, 0 }));
Assert.assertEquals(0, optimizer.getRMS(), 1e-10); Assert.assertEquals(0, optimizer.getRMS(), 1e-10);
Assert.assertEquals(3, optimum.getPoint()[0], 1e-10); Assert.assertEquals(3, optimum.getPoint()[0], 1e-10);
@ -234,7 +235,7 @@ public abstract class AbstractLeastSquaresOptimizerAbstractTest {
problem.getModelFunction(), problem.getModelFunction(),
problem.getModelFunctionJacobian(), problem.getModelFunctionJacobian(),
problem.getTarget(), problem.getTarget(),
new NonCorrelatedWeight(new double[] { 1, 1, 1 }), new Weight(new double[] { 1, 1, 1 }),
new InitialGuess(new double[] { 0, 0, 0 })); new InitialGuess(new double[] { 0, 0, 0 }));
} }
@ -252,7 +253,7 @@ public abstract class AbstractLeastSquaresOptimizerAbstractTest {
problem1.getModelFunction(), problem1.getModelFunction(),
problem1.getModelFunctionJacobian(), problem1.getModelFunctionJacobian(),
problem1.getTarget(), problem1.getTarget(),
new NonCorrelatedWeight(new double[] { 1, 1, 1, 1 }), new Weight(new double[] { 1, 1, 1, 1 }),
new InitialGuess(new double[] { 0, 1, 2, 3 })); new InitialGuess(new double[] { 0, 1, 2, 3 }));
Assert.assertEquals(0, optimizer.getRMS(), 1e-10); Assert.assertEquals(0, optimizer.getRMS(), 1e-10);
Assert.assertEquals(1, optimum1.getPoint()[0], 1e-10); Assert.assertEquals(1, optimum1.getPoint()[0], 1e-10);
@ -271,7 +272,7 @@ public abstract class AbstractLeastSquaresOptimizerAbstractTest {
problem2.getModelFunction(), problem2.getModelFunction(),
problem2.getModelFunctionJacobian(), problem2.getModelFunctionJacobian(),
problem2.getTarget(), problem2.getTarget(),
new NonCorrelatedWeight(new double[] { 1, 1, 1, 1 }), new Weight(new double[] { 1, 1, 1, 1 }),
new InitialGuess(new double[] { 0, 1, 2, 3 })); new InitialGuess(new double[] { 0, 1, 2, 3 }));
Assert.assertEquals(0, optimizer.getRMS(), 1e-10); Assert.assertEquals(0, optimizer.getRMS(), 1e-10);
Assert.assertEquals(-81, optimum2.getPoint()[0], 1e-8); Assert.assertEquals(-81, optimum2.getPoint()[0], 1e-8);
@ -294,7 +295,7 @@ public abstract class AbstractLeastSquaresOptimizerAbstractTest {
problem.getModelFunction(), problem.getModelFunction(),
problem.getModelFunctionJacobian(), problem.getModelFunctionJacobian(),
problem.getTarget(), problem.getTarget(),
new NonCorrelatedWeight(new double[] { 1, 1, 1 }), new Weight(new double[] { 1, 1, 1 }),
new InitialGuess(new double[] { 7, 6, 5, 4 })); new InitialGuess(new double[] { 7, 6, 5, 4 }));
Assert.assertEquals(0, optimizer.getRMS(), 1e-10); Assert.assertEquals(0, optimizer.getRMS(), 1e-10);
} }
@ -315,7 +316,7 @@ public abstract class AbstractLeastSquaresOptimizerAbstractTest {
problem.getModelFunction(), problem.getModelFunction(),
problem.getModelFunctionJacobian(), problem.getModelFunctionJacobian(),
problem.getTarget(), problem.getTarget(),
new NonCorrelatedWeight(new double[] { 1, 1, 1, 1, 1 }), new Weight(new double[] { 1, 1, 1, 1, 1 }),
new InitialGuess(new double[] { 2, 2, 2, 2, 2, 2 })); new InitialGuess(new double[] { 2, 2, 2, 2, 2, 2 }));
Assert.assertEquals(0, optimizer.getRMS(), 1e-10); Assert.assertEquals(0, optimizer.getRMS(), 1e-10);
Assert.assertEquals(3, optimum.getPointRef()[2], 1e-10); Assert.assertEquals(3, optimum.getPointRef()[2], 1e-10);
@ -338,7 +339,7 @@ public abstract class AbstractLeastSquaresOptimizerAbstractTest {
problem.getModelFunction(), problem.getModelFunction(),
problem.getModelFunctionJacobian(), problem.getModelFunctionJacobian(),
problem.getTarget(), problem.getTarget(),
new NonCorrelatedWeight(new double[] { 1, 1, 1 }), new Weight(new double[] { 1, 1, 1 }),
new InitialGuess(new double[] { 1, 1 })); new InitialGuess(new double[] { 1, 1 }));
Assert.assertEquals(0, optimizer.getRMS(), 1e-10); Assert.assertEquals(0, optimizer.getRMS(), 1e-10);
Assert.assertEquals(2, optimum.getPointRef()[0], 1e-10); Assert.assertEquals(2, optimum.getPointRef()[0], 1e-10);
@ -358,7 +359,7 @@ public abstract class AbstractLeastSquaresOptimizerAbstractTest {
problem.getModelFunction(), problem.getModelFunction(),
problem.getModelFunctionJacobian(), problem.getModelFunctionJacobian(),
problem.getTarget(), problem.getTarget(),
new NonCorrelatedWeight(new double[] { 1, 1, 1 }), new Weight(new double[] { 1, 1, 1 }),
new InitialGuess(new double[] { 1, 1 })); new InitialGuess(new double[] { 1, 1 }));
Assert.assertTrue(optimizer.getRMS() > 0.1); Assert.assertTrue(optimizer.getRMS() > 0.1);
} }
@ -374,7 +375,7 @@ public abstract class AbstractLeastSquaresOptimizerAbstractTest {
problem.getModelFunction(), problem.getModelFunction(),
problem.getModelFunctionJacobian(), problem.getModelFunctionJacobian(),
problem.getTarget(), problem.getTarget(),
new NonCorrelatedWeight(new double[] { 1, 1 }), new Weight(new double[] { 1, 1 }),
new InitialGuess(new double[] { 0, 0 })); new InitialGuess(new double[] { 0, 0 }));
Assert.assertEquals(0, optimizer.getRMS(), 1e-10); Assert.assertEquals(0, optimizer.getRMS(), 1e-10);
Assert.assertEquals(-1, optimum.getPoint()[0], 1e-10); Assert.assertEquals(-1, optimum.getPoint()[0], 1e-10);
@ -384,7 +385,7 @@ public abstract class AbstractLeastSquaresOptimizerAbstractTest {
problem.getModelFunction(), problem.getModelFunction(),
problem.getModelFunctionJacobian(), problem.getModelFunctionJacobian(),
problem.getTarget(), problem.getTarget(),
new NonCorrelatedWeight(new double[] { 1 }), new Weight(new double[] { 1 }),
new InitialGuess(new double[] { 0, 0 })); new InitialGuess(new double[] { 0, 0 }));
} }
@ -399,7 +400,7 @@ public abstract class AbstractLeastSquaresOptimizerAbstractTest {
problem.getModelFunction(), problem.getModelFunction(),
problem.getModelFunctionJacobian(), problem.getModelFunctionJacobian(),
problem.getTarget(), problem.getTarget(),
new NonCorrelatedWeight(new double[] { 1, 1 }), new Weight(new double[] { 1, 1 }),
new InitialGuess(new double[] { 0, 0 })); new InitialGuess(new double[] { 0, 0 }));
Assert.assertEquals(0, optimizer.getRMS(), 1e-10); Assert.assertEquals(0, optimizer.getRMS(), 1e-10);
Assert.assertEquals(-1, optimum.getPoint()[0], 1e-10); Assert.assertEquals(-1, optimum.getPoint()[0], 1e-10);
@ -409,7 +410,7 @@ public abstract class AbstractLeastSquaresOptimizerAbstractTest {
problem.getModelFunction(), problem.getModelFunction(),
problem.getModelFunctionJacobian(), problem.getModelFunctionJacobian(),
new Target(new double[] { 1 }), new Target(new double[] { 1 }),
new NonCorrelatedWeight(new double[] { 1 }), new Weight(new double[] { 1 }),
new InitialGuess(new double[] { 0, 0 })); new InitialGuess(new double[] { 0, 0 }));
} }
@ -427,7 +428,7 @@ public abstract class AbstractLeastSquaresOptimizerAbstractTest {
circle.getModelFunction(), circle.getModelFunction(),
circle.getModelFunctionJacobian(), circle.getModelFunctionJacobian(),
new Target(new double[] { 0, 0, 0, 0, 0 }), new Target(new double[] { 0, 0, 0, 0, 0 }),
new NonCorrelatedWeight(new double[] { 1, 1, 1, 1, 1 }), new Weight(new double[] { 1, 1, 1, 1, 1 }),
new InitialGuess(new double[] { 98.680, 47.345 })); new InitialGuess(new double[] { 98.680, 47.345 }));
Assert.assertTrue(optimizer.getEvaluations() < 10); Assert.assertTrue(optimizer.getEvaluations() < 10);
double rms = optimizer.getRMS(); double rms = optimizer.getRMS();
@ -455,7 +456,7 @@ public abstract class AbstractLeastSquaresOptimizerAbstractTest {
circle.getModelFunction(), circle.getModelFunction(),
circle.getModelFunctionJacobian(), circle.getModelFunctionJacobian(),
new Target(target), new Target(target),
new NonCorrelatedWeight(weights), new Weight(weights),
new InitialGuess(new double[] { 98.680, 47.345 })); new InitialGuess(new double[] { 98.680, 47.345 }));
cov = optimizer.computeCovariances(optimum.getPoint(), 1e-14); cov = optimizer.computeCovariances(optimum.getPoint(), 1e-14);
Assert.assertEquals(0.0016, cov[0][0], 0.001); Assert.assertEquals(0.0016, cov[0][0], 0.001);
@ -481,7 +482,7 @@ public abstract class AbstractLeastSquaresOptimizerAbstractTest {
circle.getModelFunction(), circle.getModelFunction(),
circle.getModelFunctionJacobian(), circle.getModelFunctionJacobian(),
new Target(target), new Target(target),
new NonCorrelatedWeight(weights), new Weight(weights),
new InitialGuess(new double[] { -12, -12 })); new InitialGuess(new double[] { -12, -12 }));
Vector2D center = new Vector2D(optimum.getPointRef()[0], optimum.getPointRef()[1]); Vector2D center = new Vector2D(optimum.getPointRef()[0], optimum.getPointRef()[1]);
Assert.assertTrue(optimizer.getEvaluations() < 25); Assert.assertTrue(optimizer.getEvaluations() < 25);
@ -508,7 +509,7 @@ public abstract class AbstractLeastSquaresOptimizerAbstractTest {
circle.getModelFunction(), circle.getModelFunction(),
circle.getModelFunctionJacobian(), circle.getModelFunctionJacobian(),
new Target(target), new Target(target),
new NonCorrelatedWeight(weights), new Weight(weights),
new InitialGuess(new double[] { 0, 0 })); new InitialGuess(new double[] { 0, 0 }));
Assert.assertEquals(-0.1517383071957963, optimum.getPointRef()[0], 1e-6); Assert.assertEquals(-0.1517383071957963, optimum.getPointRef()[0], 1e-6);
Assert.assertEquals(0.2074999736353867, optimum.getPointRef()[1], 1e-6); Assert.assertEquals(0.2074999736353867, optimum.getPointRef()[1], 1e-6);
@ -562,7 +563,7 @@ public abstract class AbstractLeastSquaresOptimizerAbstractTest {
problem.getModelFunction(), problem.getModelFunction(),
problem.getModelFunctionJacobian(), problem.getModelFunctionJacobian(),
new Target(data[1]), new Target(data[1]),
new NonCorrelatedWeight(w), new Weight(w),
new InitialGuess(initial)); new InitialGuess(initial));
final double[] actual = optimum.getPoint(); final double[] actual = optimum.getPoint();

13
src/test/java/org/apache/commons/math3/optim/nonlinear/vector/jacobian/AbstractLeastSquaresOptimizerTest.java
View File

@ -15,15 +15,14 @@ package org.apache.commons.math3.optim.nonlinear.vector.jacobian;
import java.io.IOException; import java.io.IOException;
import java.util.Arrays; import java.util.Arrays;
import org.apache.commons.math3.optim.PointVectorValuePair;
import org.apache.commons.math3.optim.InitialGuess; import org.apache.commons.math3.optim.InitialGuess;
import org.apache.commons.math3.optim.MaxEval; import org.apache.commons.math3.optim.MaxEval;
import org.apache.commons.math3.optim.PointVectorValuePair;
import org.apache.commons.math3.optim.nonlinear.vector.Target; import org.apache.commons.math3.optim.nonlinear.vector.Target;
import org.apache.commons.math3.optim.nonlinear.vector.NonCorrelatedWeight; import org.apache.commons.math3.optim.nonlinear.vector.Weight;
import org.apache.commons.math3.util.FastMath; import org.apache.commons.math3.util.FastMath;
import org.junit.Assert;
import org.junit.Test; import org.junit.Test;
import org.junit.Assert;
public class AbstractLeastSquaresOptimizerTest { public class AbstractLeastSquaresOptimizerTest {
@ -57,7 +56,7 @@ public class AbstractLeastSquaresOptimizerTest {
problem.getModelFunction(), problem.getModelFunction(),
problem.getModelFunctionJacobian(), problem.getModelFunctionJacobian(),
new Target(y), new Target(y),
new NonCorrelatedWeight(w), new Weight(w),
new InitialGuess(a)); new InitialGuess(a));
final double expected = dataset.getResidualSumOfSquares(); final double expected = dataset.getResidualSumOfSquares();
final double actual = optimizer.getChiSquare(); final double actual = optimizer.getChiSquare();
@ -82,7 +81,7 @@ public class AbstractLeastSquaresOptimizerTest {
problem.getModelFunction(), problem.getModelFunction(),
problem.getModelFunctionJacobian(), problem.getModelFunctionJacobian(),
new Target(y), new Target(y),
new NonCorrelatedWeight(w), new Weight(w),
new InitialGuess(a)); new InitialGuess(a));
final double expected = FastMath final double expected = FastMath
@ -111,7 +110,7 @@ public class AbstractLeastSquaresOptimizerTest {
problem.getModelFunction(), problem.getModelFunction(),
problem.getModelFunctionJacobian(), problem.getModelFunctionJacobian(),
new Target(y), new Target(y),
new NonCorrelatedWeight(w), new Weight(w),
new InitialGuess(a)); new InitialGuess(a));
final double[] sig = optimizer.computeSigma(optimum.getPoint(), 1e-14); final double[] sig = optimizer.computeSigma(optimum.getPoint(), 1e-14);

17
src/test/java/org/apache/commons/math3/optim/nonlinear/vector/jacobian/AbstractLeastSquaresOptimizerTestValidation.java
View File

@ -13,21 +13,20 @@
*/ */
package org.apache.commons.math3.optim.nonlinear.vector.jacobian; package org.apache.commons.math3.optim.nonlinear.vector.jacobian;
import java.awt.geom.Point2D;
import java.util.ArrayList;
import java.util.Arrays; import java.util.Arrays;
import java.util.List; import java.util.List;
import java.util.ArrayList;
import java.awt.geom.Point2D;
import org.apache.commons.math3.optim.PointVectorValuePair;
import org.apache.commons.math3.optim.InitialGuess; import org.apache.commons.math3.optim.InitialGuess;
import org.apache.commons.math3.optim.MaxEval; import org.apache.commons.math3.optim.MaxEval;
import org.apache.commons.math3.optim.PointVectorValuePair;
import org.apache.commons.math3.optim.nonlinear.vector.Target; import org.apache.commons.math3.optim.nonlinear.vector.Target;
import org.apache.commons.math3.optim.nonlinear.vector.NonCorrelatedWeight; import org.apache.commons.math3.optim.nonlinear.vector.Weight;
import org.apache.commons.math3.stat.descriptive.StatisticalSummary;
import org.apache.commons.math3.stat.descriptive.SummaryStatistics; import org.apache.commons.math3.stat.descriptive.SummaryStatistics;
import org.apache.commons.math3.stat.descriptive.StatisticalSummary;
import org.apache.commons.math3.util.FastMath; import org.apache.commons.math3.util.FastMath;
import org.junit.Assert;
import org.junit.Test; import org.junit.Test;
import org.junit.Assert;
/** /**
* This class demonstrates the main functionality of the * This class demonstrates the main functionality of the
@ -125,7 +124,7 @@ public class AbstractLeastSquaresOptimizerTestValidation {
problem.getModelFunction(), problem.getModelFunction(),
problem.getModelFunctionJacobian(), problem.getModelFunctionJacobian(),
new Target(problem.target()), new Target(problem.target()),
new NonCorrelatedWeight(problem.weight()), new Weight(problem.weight()),
new InitialGuess(init)); new InitialGuess(init));
final double[] sigma = optim.computeSigma(optimum.getPoint(), 1e-14); final double[] sigma = optim.computeSigma(optimum.getPoint(), 1e-14);
@ -306,7 +305,7 @@ public class AbstractLeastSquaresOptimizerTestValidation {
problem.getModelFunction(), problem.getModelFunction(),
problem.getModelFunctionJacobian(), problem.getModelFunctionJacobian(),
new Target(t), new Target(t),
new NonCorrelatedWeight(w), new Weight(w),
new InitialGuess(params)); new InitialGuess(params));
return optim.getChiSquare() / (t.length - params.length); return optim.getChiSquare() / (t.length - params.length);

9
src/test/java/org/apache/commons/math3/optim/nonlinear/vector/jacobian/GaussNewtonOptimizerTest.java
View File

@ -18,14 +18,15 @@
package org.apache.commons.math3.optim.nonlinear.vector.jacobian; package org.apache.commons.math3.optim.nonlinear.vector.jacobian;
import java.io.IOException; import java.io.IOException;
import org.apache.commons.math3.exception.ConvergenceException; import org.apache.commons.math3.exception.ConvergenceException;
import org.apache.commons.math3.exception.TooManyEvaluationsException; import org.apache.commons.math3.exception.TooManyEvaluationsException;
import org.apache.commons.math3.optim.SimpleVectorValueChecker;
import org.apache.commons.math3.optim.InitialGuess; import org.apache.commons.math3.optim.InitialGuess;
import org.apache.commons.math3.optim.MaxEval; import org.apache.commons.math3.optim.MaxEval;
import org.apache.commons.math3.optim.SimpleVectorValueChecker;
import org.apache.commons.math3.optim.nonlinear.vector.Target; import org.apache.commons.math3.optim.nonlinear.vector.Target;
import org.apache.commons.math3.optim.nonlinear.vector.NonCorrelatedWeight; import org.apache.commons.math3.optim.nonlinear.vector.Weight;
import org.apache.commons.math3.optim.nonlinear.vector.ModelFunction;
import org.apache.commons.math3.optim.nonlinear.vector.ModelFunctionJacobian;
import org.junit.Test; import org.junit.Test;
/** /**
@ -132,7 +133,7 @@ public class GaussNewtonOptimizerTest
circle.getModelFunction(), circle.getModelFunction(),
circle.getModelFunctionJacobian(), circle.getModelFunctionJacobian(),
new Target(new double[] { 0, 0, 0, 0, 0 }), new Target(new double[] { 0, 0, 0, 0, 0 }),
new NonCorrelatedWeight(new double[] { 1, 1, 1, 1, 1 }), new Weight(new double[] { 1, 1, 1, 1, 1 }),
new InitialGuess(new double[] { 98.680, 47.345 })); new InitialGuess(new double[] { 98.680, 47.345 }));
} }

28
src/test/java/org/apache/commons/math3/optim/nonlinear/vector/jacobian/LevenbergMarquardtOptimizerTest.java
View File

@ -17,26 +17,28 @@
package org.apache.commons.math3.optim.nonlinear.vector.jacobian; package org.apache.commons.math3.optim.nonlinear.vector.jacobian;
import java.io.Serializable;
import java.util.ArrayList; import java.util.ArrayList;
import java.util.List; import java.util.List;
import org.apache.commons.math3.optim.PointVectorValuePair;
import org.apache.commons.math3.analysis.MultivariateMatrixFunction; import org.apache.commons.math3.optim.InitialGuess;
import org.apache.commons.math3.optim.MaxEval;
import org.apache.commons.math3.optim.nonlinear.vector.Target;
import org.apache.commons.math3.optim.nonlinear.vector.Weight;
import org.apache.commons.math3.optim.nonlinear.vector.ModelFunction;
import org.apache.commons.math3.optim.nonlinear.vector.ModelFunctionJacobian;
import org.apache.commons.math3.analysis.MultivariateVectorFunction; import org.apache.commons.math3.analysis.MultivariateVectorFunction;
import org.apache.commons.math3.analysis.MultivariateMatrixFunction;
import org.apache.commons.math3.exception.ConvergenceException;
import org.apache.commons.math3.exception.DimensionMismatchException; import org.apache.commons.math3.exception.DimensionMismatchException;
import org.apache.commons.math3.exception.TooManyEvaluationsException; import org.apache.commons.math3.exception.TooManyEvaluationsException;
import org.apache.commons.math3.geometry.euclidean.twod.Vector2D; import org.apache.commons.math3.geometry.euclidean.twod.Vector2D;
import org.apache.commons.math3.linear.SingularMatrixException; import org.apache.commons.math3.linear.SingularMatrixException;
import org.apache.commons.math3.optim.InitialGuess;
import org.apache.commons.math3.optim.MaxEval;
import org.apache.commons.math3.optim.PointVectorValuePair;
import org.apache.commons.math3.optim.nonlinear.vector.ModelFunction;
import org.apache.commons.math3.optim.nonlinear.vector.ModelFunctionJacobian;
import org.apache.commons.math3.optim.nonlinear.vector.Target;
import org.apache.commons.math3.optim.nonlinear.vector.NonCorrelatedWeight;
import org.apache.commons.math3.util.FastMath; import org.apache.commons.math3.util.FastMath;
import org.apache.commons.math3.util.Precision; import org.apache.commons.math3.util.Precision;
import org.junit.Assert; import org.junit.Assert;
import org.junit.Test; import org.junit.Test;
import org.junit.Ignore;
/** /**
* <p>Some of the unit tests are re-implementations of the MINPACK <a * <p>Some of the unit tests are re-implementations of the MINPACK <a
@ -126,7 +128,7 @@ public class LevenbergMarquardtOptimizerTest
problem.getModelFunction(), problem.getModelFunction(),
problem.getModelFunctionJacobian(), problem.getModelFunctionJacobian(),
problem.getTarget(), problem.getTarget(),
new NonCorrelatedWeight(new double[] { 1, 1, 1 }), new Weight(new double[] { 1, 1, 1 }),
new InitialGuess(new double[] { 0, 0, 0 })); new InitialGuess(new double[] { 0, 0, 0 }));
Assert.assertTrue(FastMath.sqrt(optimizer.getTargetSize()) * optimizer.getRMS() > 0.6); Assert.assertTrue(FastMath.sqrt(optimizer.getTargetSize()) * optimizer.getRMS() > 0.6);
@ -172,7 +174,7 @@ public class LevenbergMarquardtOptimizerTest
problem, problem,
problemJacobian, problemJacobian,
new Target(new double[] { 0, 0, 0, 0, 0 }), new Target(new double[] { 0, 0, 0, 0, 0 }),
new NonCorrelatedWeight(new double[] { 1, 1, 1, 1, 1 }), new Weight(new double[] { 1, 1, 1, 1, 1 }),
new InitialGuess(new double[] { 98.680, 47.345 })); new InitialGuess(new double[] { 98.680, 47.345 }));
Assert.assertTrue(!shouldFail); Assert.assertTrue(!shouldFail);
} catch (DimensionMismatchException ee) { } catch (DimensionMismatchException ee) {
@ -227,7 +229,7 @@ public class LevenbergMarquardtOptimizerTest
problem.getModelFunction(), problem.getModelFunction(),
problem.getModelFunctionJacobian(), problem.getModelFunctionJacobian(),
new Target(dataPoints[1]), new Target(dataPoints[1]),
new NonCorrelatedWeight(weights), new Weight(weights),
new InitialGuess(new double[] { 10, 900, 80, 27, 225 })); new InitialGuess(new double[] { 10, 900, 80, 27, 225 }));
final double[] solution = optimum.getPoint(); final double[] solution = optimum.getPoint();
@ -291,7 +293,7 @@ public class LevenbergMarquardtOptimizerTest
circle.getModelFunction(), circle.getModelFunction(),
circle.getModelFunctionJacobian(), circle.getModelFunctionJacobian(),
new Target(circle.target()), new Target(circle.target()),
new NonCorrelatedWeight(circle.weight()), new Weight(circle.weight()),
new InitialGuess(init)); new InitialGuess(init));
final double[] paramFound = optimum.getPoint(); final double[] paramFound = optimum.getPoint();

14
src/test/java/org/apache/commons/math3/optim/nonlinear/vector/jacobian/MinpackTest.java
View File

@ -17,18 +17,18 @@
package org.apache.commons.math3.optim.nonlinear.vector.jacobian; package org.apache.commons.math3.optim.nonlinear.vector.jacobian;
import java.io.Serializable;
import java.util.Arrays; import java.util.Arrays;
import org.apache.commons.math3.analysis.MultivariateMatrixFunction;
import org.apache.commons.math3.analysis.MultivariateVectorFunction;
import org.apache.commons.math3.exception.TooManyEvaluationsException; import org.apache.commons.math3.exception.TooManyEvaluationsException;
import org.apache.commons.math3.analysis.MultivariateVectorFunction;
import org.apache.commons.math3.analysis.MultivariateMatrixFunction;
import org.apache.commons.math3.optim.PointVectorValuePair;
import org.apache.commons.math3.optim.InitialGuess; import org.apache.commons.math3.optim.InitialGuess;
import org.apache.commons.math3.optim.MaxEval; import org.apache.commons.math3.optim.MaxEval;
import org.apache.commons.math3.optim.PointVectorValuePair; import org.apache.commons.math3.optim.nonlinear.vector.Target;
import org.apache.commons.math3.optim.nonlinear.vector.Weight;
import org.apache.commons.math3.optim.nonlinear.vector.ModelFunction; import org.apache.commons.math3.optim.nonlinear.vector.ModelFunction;
import org.apache.commons.math3.optim.nonlinear.vector.ModelFunctionJacobian; import org.apache.commons.math3.optim.nonlinear.vector.ModelFunctionJacobian;
import org.apache.commons.math3.optim.nonlinear.vector.Target;
import org.apache.commons.math3.optim.nonlinear.vector.NonCorrelatedWeight;
import org.apache.commons.math3.util.FastMath; import org.apache.commons.math3.util.FastMath;
import org.junit.Assert; import org.junit.Assert;
import org.junit.Test; import org.junit.Test;
@ -512,7 +512,7 @@ public class MinpackTest {
function.getModelFunction(), function.getModelFunction(),
function.getModelFunctionJacobian(), function.getModelFunctionJacobian(),
new Target(function.getTarget()), new Target(function.getTarget()),
new NonCorrelatedWeight(function.getWeight()), new Weight(function.getWeight()),
new InitialGuess(function.getStartPoint())); new InitialGuess(function.getStartPoint()));
Assert.assertFalse(exceptionExpected); Assert.assertFalse(exceptionExpected);
function.checkTheoreticalMinCost(optimizer.getRMS()); function.checkTheoreticalMinCost(optimizer.getRMS());