org.apache.commons.math4.analysis.polynomials.PolynomialFunctionLagrangeForm Maven / Gradle / Ivy
Go to download
Show more of this group Show more artifacts with this name
Show all versions of virtdata-lib-curves4 Show documentation
Show all versions of virtdata-lib-curves4 Show documentation
Statistical sampling library for use in virtdata libraries, based
on apache commons math 4
/*
* 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.math4.analysis.polynomials;
import org.apache.commons.math4.analysis.UnivariateFunction;
import org.apache.commons.math4.exception.DimensionMismatchException;
import org.apache.commons.math4.exception.NonMonotonicSequenceException;
import org.apache.commons.math4.exception.NumberIsTooSmallException;
import org.apache.commons.math4.exception.util.LocalizedFormats;
import org.apache.commons.math4.util.FastMath;
import org.apache.commons.math4.util.MathArrays;
/**
* Implements the representation of a real polynomial function in
*
* Lagrange Form. For reference, see Introduction to Numerical
* Analysis, ISBN 038795452X, chapter 2.
*
* The approximated function should be smooth enough for Lagrange polynomial
* to work well. Otherwise, consider using splines instead.
*
* @since 1.2
*/
public class PolynomialFunctionLagrangeForm implements UnivariateFunction {
/**
* The coefficients of the polynomial, ordered by degree -- i.e.
* coefficients[0] is the constant term and coefficients[n] is the
* coefficient of x^n where n is the degree of the polynomial.
*/
private double coefficients[];
/**
* Interpolating points (abscissas).
*/
private final double x[];
/**
* Function values at interpolating points.
*/
private final double y[];
/**
* Whether the polynomial coefficients are available.
*/
private boolean coefficientsComputed;
/**
* Construct a Lagrange polynomial with the given abscissas and function
* values. The order of interpolating points are not important.
*
* The constructor makes copy of the input arrays and assigns them.
*
* @param x interpolating points
* @param y function values at interpolating points
* @throws DimensionMismatchException if the array lengths are different.
* @throws NumberIsTooSmallException if the number of points is less than 2.
* @throws NonMonotonicSequenceException
* if two abscissae have the same value.
*/
public PolynomialFunctionLagrangeForm(double x[], double y[])
throws DimensionMismatchException, NumberIsTooSmallException, NonMonotonicSequenceException {
this.x = new double[x.length];
this.y = new double[y.length];
System.arraycopy(x, 0, this.x, 0, x.length);
System.arraycopy(y, 0, this.y, 0, y.length);
coefficientsComputed = false;
if (!verifyInterpolationArray(x, y, false)) {
MathArrays.sortInPlace(this.x, this.y);
// Second check in case some abscissa is duplicated.
verifyInterpolationArray(this.x, this.y, true);
}
}
/**
* Calculate the function value at the given point.
*
* @param z Point at which the function value is to be computed.
* @return the function value.
* @throws DimensionMismatchException if {@code x} and {@code y} have
* different lengths.
* @throws NonMonotonicSequenceException
* if {@code x} is not sorted in strictly increasing order.
* @throws NumberIsTooSmallException if the size of {@code x} is less
* than 2.
*/
@Override
public double value(double z) {
return evaluateInternal(x, y, z);
}
/**
* Returns the degree of the polynomial.
*
* @return the degree of the polynomial
*/
public int degree() {
return x.length - 1;
}
/**
* Returns a copy of the interpolating points array.
*
* Changes made to the returned copy will not affect the polynomial.
*
* @return a fresh copy of the interpolating points array
*/
public double[] getInterpolatingPoints() {
double[] out = new double[x.length];
System.arraycopy(x, 0, out, 0, x.length);
return out;
}
/**
* Returns a copy of the interpolating values array.
*
* Changes made to the returned copy will not affect the polynomial.
*
* @return a fresh copy of the interpolating values array
*/
public double[] getInterpolatingValues() {
double[] out = new double[y.length];
System.arraycopy(y, 0, out, 0, y.length);
return out;
}
/**
* Returns a copy of the coefficients array.
*
* Changes made to the returned copy will not affect the polynomial.
*
* Note that coefficients computation can be ill-conditioned. Use with caution
* and only when it is necessary.
*
* @return a fresh copy of the coefficients array
*/
public double[] getCoefficients() {
if (!coefficientsComputed) {
computeCoefficients();
}
double[] out = new double[coefficients.length];
System.arraycopy(coefficients, 0, out, 0, coefficients.length);
return out;
}
/**
* Evaluate the Lagrange polynomial using
*
* Neville's Algorithm. It takes O(n^2) time.
*
* @param x Interpolating points array.
* @param y Interpolating values array.
* @param z Point at which the function value is to be computed.
* @return the function value.
* @throws DimensionMismatchException if {@code x} and {@code y} have
* different lengths.
* @throws NonMonotonicSequenceException
* if {@code x} is not sorted in strictly increasing order.
* @throws NumberIsTooSmallException if the size of {@code x} is less
* than 2.
*/
public static double evaluate(double x[], double y[], double z)
throws DimensionMismatchException, NumberIsTooSmallException, NonMonotonicSequenceException {
if (verifyInterpolationArray(x, y, false)) {
return evaluateInternal(x, y, z);
}
// Array is not sorted.
final double[] xNew = new double[x.length];
final double[] yNew = new double[y.length];
System.arraycopy(x, 0, xNew, 0, x.length);
System.arraycopy(y, 0, yNew, 0, y.length);
MathArrays.sortInPlace(xNew, yNew);
// Second check in case some abscissa is duplicated.
verifyInterpolationArray(xNew, yNew, true);
return evaluateInternal(xNew, yNew, z);
}
/**
* Evaluate the Lagrange polynomial using
*
* Neville's Algorithm. It takes O(n^2) time.
*
* @param x Interpolating points array.
* @param y Interpolating values array.
* @param z Point at which the function value is to be computed.
* @return the function value.
* @throws DimensionMismatchException if {@code x} and {@code y} have
* different lengths.
* @throws NonMonotonicSequenceException
* if {@code x} is not sorted in strictly increasing order.
* @throws NumberIsTooSmallException if the size of {@code x} is less
* than 2.
*/
private static double evaluateInternal(double x[], double y[], double z) {
int nearest = 0;
final int n = x.length;
final double[] c = new double[n];
final double[] d = new double[n];
double min_dist = Double.POSITIVE_INFINITY;
for (int i = 0; i < n; i++) {
// initialize the difference arrays
c[i] = y[i];
d[i] = y[i];
// find out the abscissa closest to z
final double dist = FastMath.abs(z - x[i]);
if (dist < min_dist) {
nearest = i;
min_dist = dist;
}
}
// initial approximation to the function value at z
double value = y[nearest];
for (int i = 1; i < n; i++) {
for (int j = 0; j < n-i; j++) {
final double tc = x[j] - z;
final double td = x[i+j] - z;
final double divider = x[j] - x[i+j];
// update the difference arrays
final double w = (c[j+1] - d[j]) / divider;
c[j] = tc * w;
d[j] = td * w;
}
// sum up the difference terms to get the final value
if (nearest < 0.5*(n-i+1)) {
value += c[nearest]; // fork down
} else {
nearest--;
value += d[nearest]; // fork up
}
}
return value;
}
/**
* Calculate the coefficients of Lagrange polynomial from the
* interpolation data. It takes O(n^2) time.
* Note that this computation can be ill-conditioned: Use with caution
* and only when it is necessary.
*/
protected void computeCoefficients() {
final int n = degree() + 1;
coefficients = new double[n];
for (int i = 0; i < n; i++) {
coefficients[i] = 0.0;
}
// c[] are the coefficients of P(x) = (x-x[0])(x-x[1])...(x-x[n-1])
final double[] c = new double[n+1];
c[0] = 1.0;
for (int i = 0; i < n; i++) {
for (int j = i; j > 0; j--) {
c[j] = c[j-1] - c[j] * x[i];
}
c[0] *= -x[i];
c[i+1] = 1;
}
final double[] tc = new double[n];
for (int i = 0; i < n; i++) {
// d = (x[i]-x[0])...(x[i]-x[i-1])(x[i]-x[i+1])...(x[i]-x[n-1])
double d = 1;
for (int j = 0; j < n; j++) {
if (i != j) {
d *= x[i] - x[j];
}
}
final double t = y[i] / d;
// Lagrange polynomial is the sum of n terms, each of which is a
// polynomial of degree n-1. tc[] are the coefficients of the i-th
// numerator Pi(x) = (x-x[0])...(x-x[i-1])(x-x[i+1])...(x-x[n-1]).
tc[n-1] = c[n]; // actually c[n] = 1
coefficients[n-1] += t * tc[n-1];
for (int j = n-2; j >= 0; j--) {
tc[j] = c[j+1] + tc[j+1] * x[i];
coefficients[j] += t * tc[j];
}
}
coefficientsComputed = true;
}
/**
* Check that the interpolation arrays are valid.
* The arrays features checked by this method are that both arrays have the
* same length and this length is at least 2.
*
* @param x Interpolating points array.
* @param y Interpolating values array.
* @param abort Whether to throw an exception if {@code x} is not sorted.
* @throws DimensionMismatchException if the array lengths are different.
* @throws NumberIsTooSmallException if the number of points is less than 2.
* @throws NonMonotonicSequenceException
* if {@code x} is not sorted in strictly increasing order and {@code abort}
* is {@code true}.
* @return {@code false} if the {@code x} is not sorted in increasing order,
* {@code true} otherwise.
* @see #evaluate(double[], double[], double)
* @see #computeCoefficients()
*/
public static boolean verifyInterpolationArray(double x[], double y[], boolean abort)
throws DimensionMismatchException, NumberIsTooSmallException, NonMonotonicSequenceException {
if (x.length != y.length) {
throw new DimensionMismatchException(x.length, y.length);
}
if (x.length < 2) {
throw new NumberIsTooSmallException(LocalizedFormats.WRONG_NUMBER_OF_POINTS, 2, x.length, true);
}
return MathArrays.checkOrder(x, MathArrays.OrderDirection.INCREASING, true, abort);
}
}