1 /*
2 * Licensed to the Apache Software Foundation (ASF) under one or more
3 * contributor license agreements. See the NOTICE file distributed with
4 * this work for additional information regarding copyright ownership.
5 * The ASF licenses this file to You under the Apache License, Version 2.0
6 * (the "License"); you may not use this file except in compliance with
7 * the License. You may obtain a copy of the License at
8 *
9 * https://www.apache.org/licenses/LICENSE-2.0
10 *
11 * Unless required by applicable law or agreed to in writing, software
12 * distributed under the License is distributed on an "AS IS" BASIS,
13 * WITHOUT WARRANTIES OR CONDITIONS OF ANY KIND, either express or implied.
14 * See the License for the specific language governing permissions and
15 * limitations under the License.
16 */
17
18 /*
19 * This is not the original file distributed by the Apache Software Foundation
20 * It has been modified by the Hipparchus project
21 */
22 package org.hipparchus.fitting;
23
24 import java.util.Collection;
25
26 import org.hipparchus.analysis.polynomials.PolynomialFunction;
27 import org.hipparchus.exception.MathRuntimeException;
28 import org.hipparchus.linear.DiagonalMatrix;
29 import org.hipparchus.optim.nonlinear.vector.leastsquares.LeastSquaresBuilder;
30 import org.hipparchus.optim.nonlinear.vector.leastsquares.LeastSquaresProblem;
31
32 /**
33 * Fits points to a {@link
34 * org.hipparchus.analysis.polynomials.PolynomialFunction.Parametric polynomial}
35 * function.
36 * <br>
37 * The size of the {@link #withStartPoint(double[]) initial guess} array defines the
38 * degree of the polynomial to be fitted.
39 * They must be sorted in increasing order of the polynomial's degree.
40 * The optimal values of the coefficients will be returned in the same order.
41 *
42 */
43 public class PolynomialCurveFitter extends AbstractCurveFitter {
44 /** Parametric function to be fitted. */
45 private static final PolynomialFunction.Parametric FUNCTION = new PolynomialFunction.Parametric();
46 /** Initial guess. */
47 private final double[] initialGuess;
48 /** Maximum number of iterations of the optimization algorithm. */
49 private final int maxIter;
50
51 /**
52 * Constructor used by the factory methods.
53 *
54 * @param initialGuess Initial guess.
55 * @param maxIter Maximum number of iterations of the optimization algorithm.
56 * @throws MathRuntimeException if {@code initialGuess} is {@code null}.
57 */
58 private PolynomialCurveFitter(double[] initialGuess, int maxIter) {
59 this.initialGuess = initialGuess.clone();
60 this.maxIter = maxIter;
61 }
62
63 /**
64 * Creates a default curve fitter.
65 * Zero will be used as initial guess for the coefficients, and the maximum
66 * number of iterations of the optimization algorithm is set to
67 * {@link Integer#MAX_VALUE}.
68 *
69 * @param degree Degree of the polynomial to be fitted.
70 * @return a curve fitter.
71 *
72 * @see #withStartPoint(double[])
73 * @see #withMaxIterations(int)
74 */
75 public static PolynomialCurveFitter create(int degree) {
76 return new PolynomialCurveFitter(new double[degree + 1], Integer.MAX_VALUE);
77 }
78
79 /**
80 * Configure the start point (initial guess).
81 * @param newStart new start point (initial guess)
82 * @return a new instance.
83 */
84 public PolynomialCurveFitter withStartPoint(double[] newStart) {
85 return new PolynomialCurveFitter(newStart.clone(),
86 maxIter);
87 }
88
89 /**
90 * Configure the maximum number of iterations.
91 * @param newMaxIter maximum number of iterations
92 * @return a new instance.
93 */
94 public PolynomialCurveFitter withMaxIterations(int newMaxIter) {
95 return new PolynomialCurveFitter(initialGuess,
96 newMaxIter);
97 }
98
99 /** {@inheritDoc} */
100 @Override
101 protected LeastSquaresProblem getProblem(Collection<WeightedObservedPoint> observations) {
102 // Prepare least-squares problem.
103 final int len = observations.size();
104 final double[] target = new double[len];
105 final double[] weights = new double[len];
106
107 int i = 0;
108 for (WeightedObservedPoint obs : observations) {
109 target[i] = obs.getY();
110 weights[i] = obs.getWeight();
111 ++i;
112 }
113
114 final AbstractCurveFitter.TheoreticalValuesFunction model =
115 new AbstractCurveFitter.TheoreticalValuesFunction(FUNCTION, observations);
116
117 if (initialGuess == null) {
118 throw MathRuntimeException.createInternalError();
119 }
120
121 // Return a new least squares problem set up to fit a polynomial curve to the
122 // observed points.
123 return new LeastSquaresBuilder().
124 maxEvaluations(Integer.MAX_VALUE).
125 maxIterations(maxIter).
126 start(initialGuess).
127 target(target).
128 weight(new DiagonalMatrix(weights)).
129 model(model.getModelFunction(), model.getModelFunctionJacobian()).
130 build();
131
132 }
133
134 }