edaTerminate {copulaedas}

Termination Methods

Description

Methods for the edaTerminate generic function.

Usage

edaTerminateMaxGen(eda, gen, fEvals, pop, popEval)
edaTerminateMaxEvals(eda, gen, fEvals, pop, popEval)
edaTerminateEval(eda, gen, fEvals, pop, popEval)
edaTerminateEvalStdDev(eda, gen, fEvals, pop, popEval)
edaTerminateCombined(...)

Arguments

eda

EDA instance.

gen

Generation.

fEvals

Evaluations of the objective function.

pop

Matrix with one row for each solution in the population.

popEval

Vector with the evaluation of each solution in pop.

...

Functions that implement termination methods.

Details

Termination methods decide when to stop the main loop of the EDA. The following termination methods are implemented.

edaTerminateMaxGen

Stop when a maximum number of generations has been reached. The parameter maxGen specifies the number of generations (default value: 100). This is the default method of the edaTerminate generic function.

edaTerminateMaxEvals

Stop when a maximum number of evaluations of the objective function has been reached. The parameter maxEvals specifies the number of evaluations (default value: 1000.)

edaTerminateEval

Stop when a given value of the objective function has been reached. The parameters fEval (default value: 0) and fEvalTol (default value: 1e-06) set the value of the objective function and the tolerance, respectively.

edaTerminateEvalStdDev

Stop when the standard deviation of the evaluation of the solutions in the population is less than the value given by the parameter fEvalStdDev (default value: 1e-02)

.

edaTerminateCombined

Evaluate all the termination criteria specified in ... and stop if (at least) one of them returns TRUE.

Value

A logical value that indicates if the algorithm should stop.

References

Gonzalez-Fernandez Y, Soto M (2014). copulaedas: An R Package for Estimation of Distribution Algorithms Based on Copulas. Journal of Statistical Software, 58(9), 1-34. http://www.jstatsoft.org/v58/i09/.