bayesopt::NLOPT_Optimization Class Reference

Collaboration diagram for bayesopt::NLOPT_Optimization:

Public Member Functions
	NLOPT_Optimization (RBOptimizable *rbo, size_t dim)
	NLOPT_Optimization (RGBOptimizable *rgbo, size_t dim)
void	setAlgorithm (innerOptAlgorithms newAlg)
	Sets the optimization algorithm.
void	setMaxEvals (size_t meval)
	Sets the maximum number of function evaluations. More...
void	setLimits (const vectord &down, const vectord &up)
	Limits of the hypercube. More...
void	setLimits (double down, double up)
	Limits of the hypercube assuming that all dimensions have the same limits. More...
double	run (vectord &Xnext)
	Launch the inner optimization algorithm. More...
double	localTrialAround (vectord &Xnext)
	Try some local optimization around a point. More...

Static Public Member Functions
static double	evaluate_nlopt (unsigned int n, const double *x, double *grad, void *my_func_data)
	Wrapper of inner optimization to be evaluated by NLOPT. More...
static double	evaluate_nlopt_grad (unsigned int n, const double *x, double *grad, void *my_func_data)
	Wrapper of inner optimization to be evaluated by NLOPT. More...

Private Member Functions
	NLOPT_Optimization (NLOPT_Optimization &copy)

Private Attributes
RBOptimizableWrapper *	rbobj
RGBOptimizableWrapper *	rgbobj
innerOptAlgorithms	alg
std::vector< double >	mDown
std::vector< double >	mUp
size_t	maxEvals

Detailed Description

Definition at line 44 of file inneroptimization.hpp.

Member Function Documentation

evaluate_nlopt()

double bayesopt::NLOPT_Optimization::evaluate_nlopt ( unsigned int  n, const double *  x, double *  grad, void *  my_func_data  )

static

Wrapper of inner optimization to be evaluated by NLOPT.

Parameters

n	# of dimensions
x	input point
grad	(NOT USED. Only for compatibily with NLOPT template, see evaluate_nlopt_grad)
my_func_data	pointer to the NLOPT_Optimization object

Returns

function evaluation

Definition at line 246 of file inneroptimization.cpp.

Referenced by run().

evaluate_nlopt_grad()

double bayesopt::NLOPT_Optimization::evaluate_nlopt_grad ( unsigned int  n, const double *  x, double *  grad, void *  my_func_data  )

static

Wrapper of inner optimization to be evaluated by NLOPT.

Parameters

n	# of dimensions
x	input point
grad	returns gradient evaluation
my_func_data	pointer to the NLOPT_Optimization object

Returns

function evaluation

Definition at line 259 of file inneroptimization.cpp.

Referenced by run().

localTrialAround()

double bayesopt::NLOPT_Optimization::localTrialAround

(

vectord &

Xnext

)

Try some local optimization around a point.

Parameters

Xnext

input: initial guess, output: result

Returns

minimum value

Definition at line 109 of file inneroptimization.cpp.

run()

double bayesopt::NLOPT_Optimization::run

(

vectord &

Xnext

)

Launch the inner optimization algorithm.

Parameters

Xnext

input: initial guess, output: result

Returns

minimum value

Definition at line 150 of file inneroptimization.cpp.

References bayesopt::BOBYQA, bayesopt::COMBINED, bayesopt::DIRECT, evaluate_nlopt(), evaluate_nlopt_grad(), and bayesopt::LBFGS.

setLimits() [1/2]

void bayesopt::NLOPT_Optimization::setLimits ( const vectord &  down, const vectord &  up  )

inline

Limits of the hypercube.

Definition at line 125 of file inneroptimization.hpp.

setLimits() [2/2]

void bayesopt::NLOPT_Optimization::setLimits ( double  down, double  up  )

inline

Limits of the hypercube assuming that all dimensions have the same limits.

Definition at line 131 of file inneroptimization.hpp.

setMaxEvals()

void bayesopt::NLOPT_Optimization::setMaxEvals ( size_t  meval )

inline

Sets the maximum number of function evaluations.

Depending on the algorithm, it might stops earlier if convergence is reached.

Definition at line 122 of file inneroptimization.hpp.

---

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

• inneroptimization.hpp
• inneroptimization.cpp