sMC Class Reference

Abstract base class that defines a structured Monte Carlo sampling algorithm. More...

#include <sMC.h>

Inheritance diagram for sMC::

List of all members.

General Information
int	write (ostream &)
	Print information about the optimizer.
int	read (istream &)
	Read information about the optimizer (not presently supported).
Uniform	urnd
Iteration Controls
virtual void	setup (REAL &fret)=0
	Calculate first point and setup.
virtual void	generate_init ()
	Initialize for another iteration.
virtual void	generate_point ()=0
	Generate next point.
virtual void	do_ls (REAL &)
	Execute local search on current iterate.
virtual void	evaluate_point (REAL &fp)=0
	Evaluate generated point.
virtual void	set_best (int ndx)=0
	Set best point equal to the current iterate.
virtual void	finish_up ()=0
	Finish up; set the function memory to the best point.
virtual void	synchronize ()=0
	Finish up; set the function memory to the best point.
int	curr_ndx
	Current point index.
Public Methods
General Information
	sMC ()
int	minimize ()
	Performs minimization. More...
void	set_rng (RNG *rng_)
	Sets the random number generator.
Configuration Controls
void	set_batch_size (const int size)
Protected Attributes
Configuration Controls
REAL	freq
	Local search frequency.
int	npartitions
	Number of partitions.
int	batch_size
	The number of evaluations per iteration.

---

Detailed Description

Abstract base class that defines a structured Monte Carlo sampling algorithm.

The default is to define a MC to uniformly sample over the entire search space. The user can also define partitions to allow this class to sample K disjoint regions of the search space in each iteration. Subclasses of \Ref{sMC} can specify the number of partitions in {\bf num\_partitions} and they can specify the frequency of local search in \Ref{freq}.

\Ref{sMC} defines the main loop for a sMC algorithm. The main loop utilizes the protected methods defined in the Iteration Controls section.

---

Member Function Documentation

int sMC::minimize (    )  [virtual]

Performs minimization. This method is defined be the subclass. The \Ref{check_convergence} method can be used to utilize the termination controls defined by this class. Also, the \Ref{debug_io} method can be used to print general debugging information in a standard format. If the developer wishes to provide additional debugging information, the \Ref{virt_debug_io} method can be redefined. This method is called from \Ref{debug_io}. Reimplemented from BaseOptimizer. Reimplemented in sMCint, and sMCreal.

---

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

• sMC.h