madness::NonlinearSolverND< NDIM > Class Template Reference

A simple Krylov-subspace nonlinear equation solver. More...

#include <nonlinsol.h>

Collaboration diagram for madness::NonlinearSolverND< NDIM >:

Public Member Functions
	NonlinearSolverND (unsigned int maxsub=10)
unsigned int	get_maxsub () const
void	set_maxsub (const unsigned int &new_maxsub)
Function< double, NDIM >	update (const Function< double, NDIM > &u, const Function< double, NDIM > &r, const double rcondtol=1e-8, const double cabsmax=1000.0)
	Computes next trial solution vector.

Public Attributes
bool	do_print

Private Attributes
unsigned int	maxsub
	Maximum size of subspace dimension.
real_tensor	Q
std::vector< Function< double, NDIM > >	rlist
std::vector< Function< double, NDIM > >	ulist

Detailed Description

template<size_t NDIM>
class madness::NonlinearSolverND< NDIM >

A simple Krylov-subspace nonlinear equation solver.

Constructor & Destructor Documentation

NonlinearSolverND()

template<size_t NDIM>

madness::NonlinearSolverND< NDIM >::NonlinearSolverND ( unsigned int  maxsub = 10 )

inline

Member Function Documentation

get_maxsub()

template<size_t NDIM>

unsigned int madness::NonlinearSolverND< NDIM >::get_maxsub ( ) const

inline

References madness::NonlinearSolverND< NDIM >::maxsub.

set_maxsub()

template<size_t NDIM>

void madness::NonlinearSolverND< NDIM >::set_maxsub ( const unsigned int &  new_maxsub )

inline

References madness::NonlinearSolverND< NDIM >::maxsub.

update()

template<size_t NDIM>

Function< double, NDIM > madness::NonlinearSolverND< NDIM >::update ( const Function< double, NDIM > &  u, const Function< double, NDIM > &  r, const double  rcondtol = 1e-8, const double  cabsmax = 1000.0  )

inline

Computes next trial solution vector.

You are responsible for performing step restriction or line search (not necessary for linear problems).

Parameters

u	Current solution vector
r	Corresponding residual

Returns

Next trial solution vector

Parameters

[in]	rcondtol	rcond less than this will cause the subspace to be shrunk due to linear dependence
[in]	cabsmax	maximum element of c greater than this will cause the subspace to be shrunk due to linear dependence

References c, madness::change_tree_state(), madness::check_linear_dependence(), madness::compressed, madness::copy(), madness::NonlinearSolverND< NDIM >::do_print, madness::Function< T, NDIM >::gaxpy(), madness::inner(), madness::KAIN(), madness::NonlinearSolverND< NDIM >::maxsub, madness::print(), madness::NonlinearSolverND< NDIM >::Q, madness::reconstructed, madness::NonlinearSolverND< NDIM >::rlist, madness::Function< T, NDIM >::size(), madness::FunctionFactory< T, NDIM >::treestate(), madness::Function< T, NDIM >::truncate(), madness::TT_FULL, u(), and madness::NonlinearSolverND< NDIM >::ulist.

Referenced by madness::CC2::iterate_pair(), and madness::MP2::solve_residual_equations().

Member Data Documentation

do_print

template<size_t NDIM>

bool madness::NonlinearSolverND< NDIM >::do_print

Referenced by madness::CC2::iterate_pair(), madness::MP2::solve_residual_equations(), and madness::NonlinearSolverND< NDIM >::update().

maxsub

template<size_t NDIM>

unsigned int madness::NonlinearSolverND< NDIM >::maxsub

private

Maximum size of subspace dimension.

Referenced by madness::NonlinearSolverND< NDIM >::get_maxsub(), madness::NonlinearSolverND< NDIM >::set_maxsub(), and madness::NonlinearSolverND< NDIM >::update().

Q

template<size_t NDIM>

real_tensor madness::NonlinearSolverND< NDIM >::Q

private

Referenced by madness::NonlinearSolverND< NDIM >::update().

rlist

template<size_t NDIM>

std::vector<Function<double,NDIM> > madness::NonlinearSolverND< NDIM >::rlist

private

Referenced by madness::NonlinearSolverND< NDIM >::update().

ulist

template<size_t NDIM>

std::vector<Function<double,NDIM> > madness::NonlinearSolverND< NDIM >::ulist

private

Referenced by madness::NonlinearSolverND< NDIM >::update().

---

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

• nonlinsol.h