Main class describing an evolutionary optimization strategy. Every algorithm in libcmaes descends from this class, and bring its functionalities to an ESOptimizer object. More...

#include <libcmaes/esostrategy.h>

Inheritance diagram for libcmaes::ESOStrategy< TParameters, TSolutions, TStopCriteria >:

Public Member Functions
	ESOStrategy ()
	dummy constructor.

	ESOStrategy (FitFunc &func, TParameters &parameters)
	constructor

	ESOStrategy (FitFunc &func, TParameters &parameters, const TSolutions &solutions)
	constructor for starting from an existing solution.

dMat	ask ()
	Generates a set of candidate points.

void	eval (const dMat &candidates, const dMat &phenocandidates=dMat(0, 0))
	Evaluates a set of candidates against the objective function. The procedure is multithreaded and stores both the candidates and their f-value into the _solutions object that bears the current set of potential solutions to the optimization problem.

void	tell ()
	Updates the state of the stochastic search, and prepares for the next iteration.

bool	stop ()
	Decides whether to stop the search for solutions.

int	optimize (const EvalFunc &evalf, const AskFunc &askf, const TellFunc &tellf)
	Finds the minimum of the objective function. It makes alternative calls to ask(), tell() and stop() until one of the termination criteria triggers.

void	inc_iter ()
	increment iteration count.

void	update_fevals (const int &evals)
	updates the consumed budget of objective function evaluations.

void	set_gradient_func (GradFunc &gfunc)
	sets the gradient function, if available.

void	set_progress_func (ProgressFunc< TParameters, TSolutions > &pfunc)
	Sets the possibly custom progress function, that is called in between every search step, and gives an outside user a simple way to witness progress of the algorithm, as well as to add custom termination criteria.

void	start_from_solution (const TSolutions &sol)
	starts optimization from a given solution object.

void	set_plot_func (PlotFunc< TParameters, TSolutions > &pffunc)
	Sets the possibly custom plot to file function, that is useful for storing into file various possibly custom variable values for each step until termination.

dVec	gradf (const dVec &x)
	returns numerical gradient of objective function at x.

dVec	gradgp (const dVec &x) const
	returns the numerical gradient of the objective function in phenotype space

double	edm ()
	computes expected distance to minimum (EDM).

TSolutions &	get_solutions ()
	returns reference to current solution object

TParameters &	get_parameters ()
	returns reference to current optimization parameters object

double	fitfunc (const double *x, const int N)
	execute objective function

void	uncertainty_handling ()
	uncertainty handling scheme that computes and uncertainty level based on a dual candidate ranking.

void	perform_uh (const dMat &candidates, const dMat &phenocandidates, int &nfcalls)
	uncertainty handling scheme that perform completely the reevaluation of solutions.

void	select_candidates_uh (const dMat &candidates, const dMat &phenocandidates, dMat &candidates_uh)
	part of the ucertainty handling scheme that select which candidates should be reevaluated.

void	eval_candidates_uh (const dMat &candidates, const dMat &candidates_uh, std::vector< RankedCandidate > &nvcandidates, int &nfcalls)
	part of the ucertainty handling scheme that evaluate the candidates to be reevaluated.

void	set_candidates_uh (const std::vector< RankedCandidate > &nvcandidates)
	part of the ucertainty handling scheme that set the results of evaluation to the solutions.

void	tpa_update ()
	updates the two-point adaptation average rank difference for the step-size adaptation mechanism

Candidate	best_solution () const

void	set_initial_elitist (const bool &e)

Protected Attributes
FitFunc	_func

int	_nevals

int	_niter

TSolutions	_solutions

TParameters	_parameters

ProgressFunc< TParameters, TSolutions >	_pfunc

GradFunc	_gfunc = nullptr

PlotFunc< TParameters, TSolutions >	_pffunc

FitFunc	_funcaux

bool	_initial_elitist = false

Private Attributes
std::mt19937	_uhgen

std::uniform_real_distribution	_uhunif

Eigen::EigenMultivariateNormal< double >	_uhesolver

Detailed Description

template<class TParameters, class TSolutions, class TStopCriteria>
class libcmaes::ESOStrategy< TParameters, TSolutions, TStopCriteria >

Main class describing an evolutionary optimization strategy. Every algorithm in libcmaes descends from this class, and bring its functionalities to an ESOptimizer object.

Constructor & Destructor Documentation

◆ ESOStrategy() [1/2]

template<class TParameters , class TSolutions , class TStopCriteria >

libcmaes::ESOStrategy< TParameters, TSolutions, TStopCriteria >::ESOStrategy	(	FitFunc &	func,
		TParameters &	parameters
	)

constructor

Parameters

func	function to minimize
parameters	optimization parameters

◆ ESOStrategy() [2/2]

template<class TParameters , class TSolutions , class TStopCriteria >

libcmaes::ESOStrategy< TParameters, TSolutions, TStopCriteria >::ESOStrategy	(	FitFunc &	func,
		TParameters &	parameters,
		const TSolutions &	solutions
	)

constructor for starting from an existing solution.

Parameters

func	objective function to minimize
parameters	stochastic search parameters
solution	solution object to start from

Member Function Documentation

◆ ask()

template<class TParameters , class TSolutions , class TStopCriteria >

dMat libcmaes::ESOStrategy< TParameters, TSolutions, TStopCriteria >::ask ( )

Generates a set of candidate points.

Returns: A matrix whose rows contain the candidate points.

◆ edm()

template<class TParameters , class TSolutions , class TStopCriteria >

double libcmaes::ESOStrategy< TParameters, TSolutions, TStopCriteria >::edm ( )

computes expected distance to minimum (EDM).

Returns: EDM

◆ eval()

template<class TParameters , class TSolutions , class TStopCriteria >

void libcmaes::ESOStrategy< TParameters, TSolutions, TStopCriteria >::eval	(	const dMat &	candidates,
		const dMat &	phenocandidates = `dMat(0,0)`
	)

Evaluates a set of candidates against the objective function. The procedure is multithreaded and stores both the candidates and their f-value into the _solutions object that bears the current set of potential solutions to the optimization problem.

Parameters

candidates	A matrix whose rows contain the candidates.
phenocandidates	The candidates transformed into phenotype, leave empty if no pheno transform.

◆ fitfunc()

template<class TParameters , class TSolutions , class TStopCriteria >

double libcmaes::ESOStrategy< TParameters, TSolutions, TStopCriteria >::fitfunc	(	const double *	x,
		const int	N
	)

inline

execute objective function

Parameters

x	point at which to execute the function
N	dimension of array x

Returns: objective function value at x

◆ get_parameters()

template<class TParameters , class TSolutions , class TStopCriteria >

TParameters & libcmaes::ESOStrategy< TParameters, TSolutions, TStopCriteria >::get_parameters ( )

inline

returns reference to current optimization parameters object

Returns: current optimization parameters object

◆ get_solutions()

template<class TParameters , class TSolutions , class TStopCriteria >

TSolutions & libcmaes::ESOStrategy< TParameters, TSolutions, TStopCriteria >::get_solutions ( )

inline

returns reference to current solution object

Returns: current solution object

◆ gradf()

template<class TParameters , class TSolutions , class TStopCriteria >

dVec libcmaes::ESOStrategy< TParameters, TSolutions, TStopCriteria >::gradf ( const dVec & x )

returns numerical gradient of objective function at x.

Parameters

x	point at which to compute the gradient

Returns: vector of numerical gradient of the objective function at x.

◆ gradgp()

template<class TParameters , class TSolutions , class TStopCriteria >

dVec libcmaes::ESOStrategy< TParameters, TSolutions, TStopCriteria >::gradgp ( const dVec & x ) const

returns the numerical gradient of the objective function in phenotype space

Parameters

x	point in genotype coordinates at which to compute the gradient

Returns: vector of numerical gradient computed in phenotype space

◆ optimize()

template<class TParameters , class TSolutions , class TStopCriteria >

int libcmaes::ESOStrategy< TParameters, TSolutions, TStopCriteria >::optimize	(	const EvalFunc &	evalf,
		const AskFunc &	askf,
		const TellFunc &	tellf
	)

Finds the minimum of the objective function. It makes alternative calls to ask(), tell() and stop() until one of the termination criteria triggers.

Parameters

evalf	custom eval function
askf	custom ask function
tellf	custom tell function

Returns: success or error code, as defined in opti_err.h

◆ set_gradient_func()

template<class TParameters , class TSolutions , class TStopCriteria >

void libcmaes::ESOStrategy< TParameters, TSolutions, TStopCriteria >::set_gradient_func ( GradFunc & gfunc )

inline

sets the gradient function, if available.

Parameters

gfunc gradient function

◆ set_plot_func()

template<class TParameters , class TSolutions , class TStopCriteria >

void libcmaes::ESOStrategy< TParameters, TSolutions, TStopCriteria >::set_plot_func ( PlotFunc< TParameters, TSolutions > & pffunc )

inline

Sets the possibly custom plot to file function, that is useful for storing into file various possibly custom variable values for each step until termination.

Parameters

pffunc a stream to file output function

◆ set_progress_func()

template<class TParameters , class TSolutions , class TStopCriteria >

void libcmaes::ESOStrategy< TParameters, TSolutions, TStopCriteria >::set_progress_func ( ProgressFunc< TParameters, TSolutions > & pfunc )

inline

Sets the possibly custom progress function, that is called in between every search step, and gives an outside user a simple way to witness progress of the algorithm, as well as to add custom termination criteria.

Parameters

pfunc a progress function