mirror of
https://github.com/Relintai/pmlpp.git
synced 2025-01-09 17:39:37 +01:00
296 lines
10 KiB
C++
296 lines
10 KiB
C++
//
|
|
// NumericalAnalysis.cpp
|
|
//
|
|
// Created by Marc Melikyan on 11/13/20.
|
|
//
|
|
|
|
#include "numerical_analysis.h"
|
|
#include "../lin_alg/lin_alg.h"
|
|
|
|
#include <climits>
|
|
#include <cmath>
|
|
#include <iostream>
|
|
#include <string>
|
|
|
|
|
|
|
|
double MLPPNumericalAnalysis::numDiff(double (*function)(double), double x) {
|
|
double eps = 1e-10;
|
|
return (function(x + eps) - function(x)) / eps; // This is just the formal def. of the derivative.
|
|
}
|
|
|
|
double MLPPNumericalAnalysis::numDiff_2(double (*function)(double), double x) {
|
|
double eps = 1e-5;
|
|
return (function(x + 2 * eps) - 2 * function(x + eps) + function(x)) / (eps * eps);
|
|
}
|
|
|
|
double MLPPNumericalAnalysis::numDiff_3(double (*function)(double), double x) {
|
|
double eps = 1e-5;
|
|
double t1 = function(x + 3 * eps) - 2 * function(x + 2 * eps) + function(x + eps);
|
|
double t2 = function(x + 2 * eps) - 2 * function(x + eps) + function(x);
|
|
return (t1 - t2) / (eps * eps * eps);
|
|
}
|
|
|
|
double MLPPNumericalAnalysis::constantApproximation(double (*function)(double), double c) {
|
|
return function(c);
|
|
}
|
|
|
|
double MLPPNumericalAnalysis::linearApproximation(double (*function)(double), double c, double x) {
|
|
return constantApproximation(function, c) + numDiff(function, c) * (x - c);
|
|
}
|
|
|
|
double MLPPNumericalAnalysis::quadraticApproximation(double (*function)(double), double c, double x) {
|
|
return linearApproximation(function, c, x) + 0.5 * numDiff_2(function, c) * (x - c) * (x - c);
|
|
}
|
|
|
|
double MLPPNumericalAnalysis::cubicApproximation(double (*function)(double), double c, double x) {
|
|
return quadraticApproximation(function, c, x) + (1 / 6) * numDiff_3(function, c) * (x - c) * (x - c) * (x - c);
|
|
}
|
|
|
|
double MLPPNumericalAnalysis::numDiff(double (*function)(std::vector<double>), std::vector<double> x, int axis) {
|
|
// For multivariable function analysis.
|
|
// This will be used for calculating Jacobian vectors.
|
|
// Diffrentiate with respect to indicated axis. (0, 1, 2 ...)
|
|
double eps = 1e-10;
|
|
std::vector<double> x_eps = x;
|
|
x_eps[axis] += eps;
|
|
|
|
return (function(x_eps) - function(x)) / eps;
|
|
}
|
|
|
|
double MLPPNumericalAnalysis::numDiff_2(double (*function)(std::vector<double>), std::vector<double> x, int axis1, int axis2) {
|
|
//For Hessians.
|
|
double eps = 1e-5;
|
|
|
|
std::vector<double> x_pp = x;
|
|
x_pp[axis1] += eps;
|
|
x_pp[axis2] += eps;
|
|
|
|
std::vector<double> x_np = x;
|
|
x_np[axis2] += eps;
|
|
|
|
std::vector<double> x_pn = x;
|
|
x_pn[axis1] += eps;
|
|
|
|
return (function(x_pp) - function(x_np) - function(x_pn) + function(x)) / (eps * eps);
|
|
}
|
|
|
|
double MLPPNumericalAnalysis::numDiff_3(double (*function)(std::vector<double>), std::vector<double> x, int axis1, int axis2, int axis3) {
|
|
// For third order derivative tensors.
|
|
// NOTE: Approximations do not appear to be accurate for sinusodial functions...
|
|
// Should revisit this later.
|
|
double eps = INT_MAX;
|
|
|
|
std::vector<double> x_ppp = x;
|
|
x_ppp[axis1] += eps;
|
|
x_ppp[axis2] += eps;
|
|
x_ppp[axis3] += eps;
|
|
|
|
std::vector<double> x_npp = x;
|
|
x_npp[axis2] += eps;
|
|
x_npp[axis3] += eps;
|
|
|
|
std::vector<double> x_pnp = x;
|
|
x_pnp[axis1] += eps;
|
|
x_pnp[axis3] += eps;
|
|
|
|
std::vector<double> x_nnp = x;
|
|
x_nnp[axis3] += eps;
|
|
|
|
std::vector<double> x_ppn = x;
|
|
x_ppn[axis1] += eps;
|
|
x_ppn[axis2] += eps;
|
|
|
|
std::vector<double> x_npn = x;
|
|
x_npn[axis2] += eps;
|
|
|
|
std::vector<double> x_pnn = x;
|
|
x_pnn[axis1] += eps;
|
|
|
|
double thirdAxis = function(x_ppp) - function(x_npp) - function(x_pnp) + function(x_nnp);
|
|
double noThirdAxis = function(x_ppn) - function(x_npn) - function(x_pnn) + function(x);
|
|
return (thirdAxis - noThirdAxis) / (eps * eps * eps);
|
|
}
|
|
|
|
double MLPPNumericalAnalysis::newtonRaphsonMethod(double (*function)(double), double x_0, double epoch_num) {
|
|
double x = x_0;
|
|
for (int i = 0; i < epoch_num; i++) {
|
|
x -= function(x) / numDiff(function, x);
|
|
}
|
|
return x;
|
|
}
|
|
|
|
double MLPPNumericalAnalysis::halleyMethod(double (*function)(double), double x_0, double epoch_num) {
|
|
double x = x_0;
|
|
for (int i = 0; i < epoch_num; i++) {
|
|
x -= ((2 * function(x) * numDiff(function, x)) / (2 * numDiff(function, x) * numDiff(function, x) - function(x) * numDiff_2(function, x)));
|
|
}
|
|
return x;
|
|
}
|
|
|
|
double MLPPNumericalAnalysis::invQuadraticInterpolation(double (*function)(double), std::vector<double> x_0, double epoch_num) {
|
|
double x = 0;
|
|
std::vector<double> currentThree = x_0;
|
|
for (int i = 0; i < epoch_num; i++) {
|
|
double t1 = ((function(currentThree[1]) * function(currentThree[2])) / ((function(currentThree[0]) - function(currentThree[1])) * (function(currentThree[0]) - function(currentThree[2])))) * currentThree[0];
|
|
double t2 = ((function(currentThree[0]) * function(currentThree[2])) / ((function(currentThree[1]) - function(currentThree[0])) * (function(currentThree[1]) - function(currentThree[2])))) * currentThree[1];
|
|
double t3 = ((function(currentThree[0]) * function(currentThree[1])) / ((function(currentThree[2]) - function(currentThree[0])) * (function(currentThree[2]) - function(currentThree[1])))) * currentThree[2];
|
|
x = t1 + t2 + t3;
|
|
|
|
currentThree.erase(currentThree.begin());
|
|
currentThree.push_back(x);
|
|
}
|
|
return x;
|
|
}
|
|
|
|
double MLPPNumericalAnalysis::eulerianMethod(double (*derivative)(double), std::vector<double> q_0, double p, double h) {
|
|
double max_epoch = (p - q_0[0]) / h;
|
|
double x = q_0[0];
|
|
double y = q_0[1];
|
|
for (int i = 0; i < max_epoch; i++) {
|
|
y = y + h * derivative(x);
|
|
x += h;
|
|
}
|
|
return y;
|
|
}
|
|
|
|
double MLPPNumericalAnalysis::eulerianMethod(double (*derivative)(std::vector<double>), std::vector<double> q_0, double p, double h) {
|
|
double max_epoch = (p - q_0[0]) / h;
|
|
double x = q_0[0];
|
|
double y = q_0[1];
|
|
for (int i = 0; i < max_epoch; i++) {
|
|
y = y + h * derivative({ x, y });
|
|
x += h;
|
|
}
|
|
return y;
|
|
}
|
|
|
|
double MLPPNumericalAnalysis::growthMethod(double C, double k, double t) {
|
|
/*
|
|
dP/dt = kP
|
|
dP/P = kdt
|
|
integral(1/P)dP = integral(k) dt
|
|
ln|P| = kt + C_initial
|
|
|P| = e^(kt + C_initial)
|
|
|P| = e^(C_initial) * e^(kt)
|
|
P = +/- e^(C_initial) * e^(kt)
|
|
P = C * e^(kt)
|
|
*/
|
|
|
|
// auto growthFunction = [&C, &k](double t) { return C * exp(k * t); };
|
|
return C * std::exp(k * t);
|
|
}
|
|
|
|
std::vector<double> MLPPNumericalAnalysis::jacobian(double (*function)(std::vector<double>), std::vector<double> x) {
|
|
std::vector<double> jacobian;
|
|
jacobian.resize(x.size());
|
|
for (int i = 0; i < jacobian.size(); i++) {
|
|
jacobian[i] = numDiff(function, x, i); // Derivative w.r.t axis i evaluated at x. For all x_i.
|
|
}
|
|
return jacobian;
|
|
}
|
|
std::vector<std::vector<double>> MLPPNumericalAnalysis::hessian(double (*function)(std::vector<double>), std::vector<double> x) {
|
|
std::vector<std::vector<double>> hessian;
|
|
hessian.resize(x.size());
|
|
for (int i = 0; i < hessian.size(); i++) {
|
|
hessian[i].resize(x.size());
|
|
}
|
|
for (int i = 0; i < hessian.size(); i++) {
|
|
for (int j = 0; j < hessian[i].size(); j++) {
|
|
hessian[i][j] = numDiff_2(function, x, i, j);
|
|
}
|
|
}
|
|
return hessian;
|
|
}
|
|
|
|
std::vector<std::vector<std::vector<double>>> MLPPNumericalAnalysis::thirdOrderTensor(double (*function)(std::vector<double>), std::vector<double> x) {
|
|
std::vector<std::vector<std::vector<double>>> tensor;
|
|
tensor.resize(x.size());
|
|
for (int i = 0; i < tensor.size(); i++) {
|
|
tensor[i].resize(x.size());
|
|
for (int j = 0; j < tensor[i].size(); j++) {
|
|
tensor[i][j].resize(x.size());
|
|
}
|
|
}
|
|
for (int i = 0; i < tensor.size(); i++) { // O(n^3) time complexity :(
|
|
for (int j = 0; j < tensor[i].size(); j++) {
|
|
for (int k = 0; k < tensor[i][j].size(); k++)
|
|
tensor[i][j][k] = numDiff_3(function, x, i, j, k);
|
|
}
|
|
}
|
|
return tensor;
|
|
}
|
|
|
|
double MLPPNumericalAnalysis::constantApproximation(double (*function)(std::vector<double>), std::vector<double> c) {
|
|
return function(c);
|
|
}
|
|
|
|
double MLPPNumericalAnalysis::linearApproximation(double (*function)(std::vector<double>), std::vector<double> c, std::vector<double> x) {
|
|
MLPPLinAlg alg;
|
|
return constantApproximation(function, c) + alg.matmult(alg.transpose({ jacobian(function, c) }), { alg.subtraction(x, c) })[0][0];
|
|
}
|
|
|
|
double MLPPNumericalAnalysis::quadraticApproximation(double (*function)(std::vector<double>), std::vector<double> c, std::vector<double> x) {
|
|
MLPPLinAlg alg;
|
|
return linearApproximation(function, c, x) + 0.5 * alg.matmult({ (alg.subtraction(x, c)) }, alg.matmult(hessian(function, c), alg.transpose({ alg.subtraction(x, c) })))[0][0];
|
|
}
|
|
|
|
double MLPPNumericalAnalysis::cubicApproximation(double (*function)(std::vector<double>), std::vector<double> c, std::vector<double> x) {
|
|
/*
|
|
Not completely sure as the literature seldom discusses the third order taylor approximation,
|
|
in particular for multivariate cases, but ostensibly, the matrix/tensor/vector multiplies
|
|
should look something like this:
|
|
|
|
(N x N x N) (N x 1) [tensor vector mult] => (N x N x 1) => (N x N)
|
|
Perform remaining multiplies as done for the 2nd order approximation.
|
|
Result is a scalar.
|
|
*/
|
|
MLPPLinAlg alg;
|
|
std::vector<std::vector<double>> resultMat = alg.tensor_vec_mult(thirdOrderTensor(function, c), alg.subtraction(x, c));
|
|
double resultScalar = alg.matmult({ (alg.subtraction(x, c)) }, alg.matmult(resultMat, alg.transpose({ alg.subtraction(x, c) })))[0][0];
|
|
|
|
return quadraticApproximation(function, c, x) + (1 / 6) * resultScalar;
|
|
}
|
|
|
|
double MLPPNumericalAnalysis::laplacian(double (*function)(std::vector<double>), std::vector<double> x) {
|
|
MLPPLinAlg alg;
|
|
std::vector<std::vector<double>> hessian_matrix = hessian(function, x);
|
|
double laplacian = 0;
|
|
for (int i = 0; i < hessian_matrix.size(); i++) {
|
|
laplacian += hessian_matrix[i][i]; // homogenous 2nd derivs w.r.t i, then i
|
|
}
|
|
return laplacian;
|
|
}
|
|
|
|
std::string MLPPNumericalAnalysis::secondPartialDerivativeTest(double (*function)(std::vector<double>), std::vector<double> x) {
|
|
MLPPLinAlg alg;
|
|
std::vector<std::vector<double>> hessianMatrix = hessian(function, x);
|
|
/*
|
|
The reason we do this is because the 2nd partial derivative test is less conclusive for functions of variables greater than
|
|
2, and the calculations specific to the bivariate case are less computationally intensive.
|
|
*/
|
|
if (x.size() == 2) {
|
|
double det = alg.det(hessianMatrix, hessianMatrix.size());
|
|
double secondDerivative = numDiff_2(function, x, 0, 0);
|
|
if (secondDerivative > 0 && det > 0) {
|
|
return "min";
|
|
} else if (secondDerivative < 0 && det > 0) {
|
|
return "max";
|
|
} else if (det < 0) {
|
|
return "saddle";
|
|
} else {
|
|
return "test was inconclusive";
|
|
}
|
|
} else {
|
|
if (alg.positiveDefiniteChecker(hessianMatrix)) {
|
|
return "min";
|
|
} else if (alg.negativeDefiniteChecker(hessianMatrix)) {
|
|
return "max";
|
|
} else if (!alg.zeroEigenvalue(hessianMatrix)) {
|
|
return "saddle";
|
|
} else {
|
|
return "test was inconclusive";
|
|
}
|
|
}
|
|
}
|