README.md

ExpDE

Build Status CRAN_Status_Badge CRAN Downloads

ExpDE is a modular implementation of the Differential Evolution metaheuristic, which aims at providing a platform for the experimental investigation of the effect of different recombination / mutation / selection operators.

Installation

Install the package directly from CRAN using:

install.packages("ExpDE")

To get the latest version from Github, use:

# install.packages("devtools")
devtools::install_github("fcampelo/ExpDE")

How to use

Full usage instructions and examples can be found in the documentation of ExpDE::ExpDE(). Type ?ExpDE to check it.

Available Operators

Differential mutation

It is also possible to run the algorithm without using any differential mutation operator (mutation_none). Run mutation_operators() for a list of available variants.

Recombination

It is also possible to run the algorithm without using any recombination operator (recombination_none). Run recombination_operators() for a list of available variants.

Selection

Stop criteria

Example: DE/best/1/sbx on the shifted sphere problem for N = 10

popsize  <- 200
mutpars  <- list(name = "mutation_best", f = 0.6, nvecs = 1)
recpars  <- list(name = "recombination_sbx", eta = 10)
selpars  <- list(name = "selection_standard")
stopcrit <- list(names = "stop_maxeval", maxevals = 100000)
probpars <- list(name  = "sphere", xmin = rep(2, 10), xmax = rep(20, 10))
seed     <- 1234
showpars <- list(show.iters = "numbers", showevery = 10)
out      <- ExpDE(popsize, mutpars, recpars, selpars, 
                  stopcrit, probpars, seed, showpars)

If you find any bugs or has any suggestions for improvement, please feel free to contact me.

Cheers, Felipe



fcampelo/ExpDE documentation built on May 16, 2019, 12:04 p.m.