vnl_conjugate_gradient Class Reference

#include <vnl_conjugate_gradient.h>

Inheritance diagram for vnl_conjugate_gradient:

[legend]Collaboration diagram for vnl_conjugate_gradient:

[legend]List of all members.

Public Methods
	vnl_conjugate_gradient (vnl_cost_function &f)
	vnl_conjugate_gradient (vnl_cost_function &f, vnl_vector< double > &x)
void	init (vnl_cost_function &f)
	~vnl_conjugate_gradient ()
void	diagnose_outcome (vcl_ostream &) const
void	diagnose_outcome () const
bool	minimize (vnl_vector< double > &x)
Static Protected Methods
double	valuecomputer_ (double *x)
int	gradientcomputer_ (double *g, double *x)
int	valueandgradientcomputer_ (double *v, double *g, double *x)
int	preconditioner_ (double *out, double *in)
Protected Attributes
vnl_cost_function *	f_
double	final_step_size_
Friends
class	vnl_conjugate_gradient_Activate

---

Detailed Description

vnl_conjugate_gradient - real function minimization. vnl_conjugate_gradient - real function minimization

Definition at line 28 of file vnl_conjugate_gradient.h.

---

Constructor & Destructor Documentation

vnl_conjugate_gradient::vnl_conjugate_gradient (  vnl_cost_function &    f )  [inline]

Initialize with the function object that is to be minimized. Definition at line 34 of file vnl_conjugate_gradient.h. References init(), and minimize().

vnl_conjugate_gradient::vnl_conjugate_gradient (  vnl_cost_function &    f, vnl_vector< double > &    x )  [inline]

Initialize as above, and then run minimization. Definition at line 38 of file vnl_conjugate_gradient.h. References f_, final_step_size_, and vnl_conjugate_gradient_Activate.

vnl_conjugate_gradient::~vnl_conjugate_gradient (    )

Destructor.

---

Member Function Documentation

void vnl_conjugate_gradient::diagnose_outcome (    )  const

void vnl_conjugate_gradient::diagnose_outcome (  vcl_ostream &    )  const

int vnl_conjugate_gradient::gradientcomputer_ (  double *    g, double *    x )  [static, protected]

void vnl_conjugate_gradient::init (  vnl_cost_function &    f )

Initialize all variables. Referenced by vnl_conjugate_gradient().

bool vnl_conjugate_gradient::minimize (  vnl_vector< double > &    x )

Minimize the function supplied in the constructor until convergence. or failure. On return, x is such that f(x) is the lowest value achieved. Returns true for convergence, false for failure. Referenced by vnl_conjugate_gradient().

int vnl_conjugate_gradient::preconditioner_ (  double *    out, double *    in )  [static, protected]

int vnl_conjugate_gradient::valueandgradientcomputer_ (  double *    v, double *    g, double *    x )  [static, protected]

double vnl_conjugate_gradient::valuecomputer_ (  double *    x )  [static, protected]

---

Friends And Related Function Documentation

friend class vnl_conjugate_gradient_Activate [friend]

Definition at line 72 of file vnl_conjugate_gradient.h. Referenced by vnl_conjugate_gradient().

---

Member Data Documentation

vnl_cost_function* vnl_conjugate_gradient::f_ [protected]

Definition at line 67 of file vnl_conjugate_gradient.h. Referenced by vnl_conjugate_gradient().

double vnl_conjugate_gradient::final_step_size_ [protected]

Definition at line 68 of file vnl_conjugate_gradient.h. Referenced by vnl_conjugate_gradient().

---

The documentation for this class was generated from the following file:

• vnl_conjugate_gradient.h

---