Differential Evolution. More...

Inheritance diagram for DE_Optimizer:

Collaboration diagram for DE_Optimizer:

Public Member Functions
	DE_Optimizer (EvalFuncPtr fPtr=NULL, unsigned int dn=PROBLEM_DIMENSIONS, unsigned int sn=NUM_SETS, unsigned int pn=NUM_SOLUTIONS)
	see IOptimizer::IOptimizer
virtual bool	init ()
	Initialize the optimizer.
void	setF (float f)
	set parameter F
void	setCR (float cr)
	set crossover rate
void	setMutation (DE_MUTATION m)
	set mutation type
void	setCrossover (DE_CROSSOVER x)
Protected Member Functions
virtual void	initSolutions (dim3, dim3)
	Initialize solutions.
virtual void	fitnessEvaluation (dim3, dim3, bool first=false)
	Evaluate solutions.
virtual void	step (dim3, dim3)
	Generate solutions for subsequent fitnessEvaluation.
virtual void	update (dim3, dim3, bool first=false)
	Update population according to the algorithm and the results of fitnessEvaluation.
virtual void	findBest ()
	Updates global best fitnesses, indexes in SolutionSet device memory.

Detailed Description

Differential Evolution.

Definition at line 46 of file DE_Optimizer.h.

Member Function Documentation

virtual void DE_Optimizer::fitnessEvaluation	(	dim3	initializationGrid,
		dim3	initializationBlock,
		bool	first = `false`
	)		`[protected, virtual]`

Evaluate solutions.

Parameters:

initializationGrid	organization of the CUDA blocks
initializationBlock	organization of the CUDA threads
first	it signals if it is the first evaluation of the optimization

Implements IOptimizer.

virtual void DE_Optimizer::initSolutions	(	dim3	initializationGrid,
		dim3	initializationBlock
	)		`[protected, virtual]`

Initialize solutions.

Parameters:

initializationGrid	organization of the CUDA blocks
initializationBlock	organization of the CUDA threads

Implements IOptimizer.

void DE_Optimizer::setCrossover ( DE_CROSSOVER x ) [inline]

set crossover type

Possible values are:
DE_BINOMIAL
DE_CROSSOVER

Definition at line 90 of file DE_Optimizer.h.

void DE_Optimizer::setMutation ( DE_MUTATION m ) [inline]

set mutation type

Possible values are:
DE_RANDOM
DE_BEST
DE_TARGET_TO_BEST

Definition at line 83 of file DE_Optimizer.h.

virtual void DE_Optimizer::step	(	dim3	initializationGrid,
		dim3	initializationBlock
	)		`[protected, virtual]`

Generate solutions for subsequent fitnessEvaluation.

Parameters:

initializationGrid	organization of the CUDA blocks
initializationBlock	organization of the CUDA threads

Implements IOptimizer.

virtual void DE_Optimizer::update	(	dim3	initializationGrid,
		dim3	initializationBlock,
		bool	first = `false`
	)		`[protected, virtual]`

Update population according to the algorithm and the results of fitnessEvaluation.

Parameters:

initializationGrid	organization of the CUDA blocks
initializationBlock	organization of the CUDA threads
first	it signals if it is the first evaluation of the optimization

Implements IOptimizer.

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