Policy class to solve a Krylov subspace problem to get the lowest-eigenvalue eigenvector. More...

#include <solver_policies.h>

Collaboration diagram for syten::Krylov::SolverPolicies::Eigen:

Public Member Functions
SRDef	sp_convergence (DenseTensor< 2, SDef > mat, SRDef &evalue)
	Tests convergence by calculating the overlap between the last-iteration result and the current result. More...

SRDef	sp_convergence (DenseTensor< 2, SDef > mat, SRDef &evalue, StandardImmediateSolve)
	Overload. More...

template<typename Vec >
auto	sp_result (std::vector< Vec > const &krylov, SRDef, SRDef &) const
	Calculates the result based on coefficients previously-calculated in sp_convergence(). More...

template<typename Vec >
auto	sp_result (std::vector< Vec > const &krylov, SRDef, SRDef &, StandardImmediateSolve) const
	Calculates the result based on coefficients previously-calculated in sp_convergence(). More...

template<typename... Unused>
void	sp_wrapup (Unused &&...)
	Unused dummy implementation. More...

Static Private Member Functions
static DenseTensor< 1, SDef >	get_coeffs (DenseTensor< 2, SDef > mat)
	Calculates coefficients by solving the subproblem. More...

Private Attributes
DenseTensor< 1, SDef >	memoizedCoeffs
	Coefficients calculated during previous call. More...

SRDef	memoizedNorm {NAN}
	Norm of previous solution. More...

Detailed Description

Policy class to solve a Krylov subspace problem to get the lowest-eigenvalue eigenvector.

The only parameter is a reference to the place where the eigenvalue is to be stored.

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

Public Member Functions