1
2
3
4
5
6
7
8
9
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
42
43
44
|
// CppNumericalSolver
#include <iostream>
#include <Eigen/LU>
#include "isolver.h"
#include "../linesearch/armijo.h"
#ifndef NEWTONDESCENTSOLVER_H_
#define NEWTONDESCENTSOLVER_H_
namespace cppoptlib {
template<typename ProblemType>
class NewtonDescentSolver : public ISolver<ProblemType, 2> {
public:
using Superclass = ISolver<ProblemType, 2>;
using typename Superclass::Scalar;
using typename Superclass::TVector;
using typename Superclass::THessian;
void minimize(ProblemType &objFunc, TVector &x0) {
const int DIM = x0.rows();
TVector grad = TVector::Zero(DIM);
THessian hessian = THessian::Zero(DIM, DIM);
Scalar gradNorm = 0;
this->m_current.reset();
do {
objFunc.gradient(x0, grad);
objFunc.hessian(x0, hessian);
hessian += (1e-5) * THessian::Identity(DIM, DIM);
TVector delta_x = hessian.lu().solve(-grad);
const double rate = Armijo<ProblemType, 1>::linesearch(x0, delta_x, objFunc) ;
x0 = x0 + rate * delta_x;
// std::cout << "iter: "<<iter<< ", f = " << objFunc.value(x0) << ", ||g||_inf "<<gradNorm << std::endl;
++this->m_current.iterations;
this->m_current.gradNorm = grad.template lpNorm<Eigen::Infinity>();
this->m_status = checkConvergence(this->m_stop, this->m_current);
} while (objFunc.callback(this->m_current, x0) && (this->m_status == Status::Continue));
}
};
}
/* namespace cppoptlib */
#endif /* NEWTONDESCENTSOLVER_H_ */
|
