kerschke / ecr

Evolutionary Computing in R.

ecr: Evolutionary Computing in R

The ecr package provides a powerful framework for both single- and multi-objective evolutionary algorithms in R. It has build-in support for several standard genotypes like real-valued vectors, binary strings and permutations. Beside, different pre-defined building blocks in form of evolutionary operators (selectors, mutators, recombinators), e.g., k-Tournament-Selection, Gauss-Mutation, Crossover and many helper functions frequently needed in evolutionary algorithms are provided. The possibility to extend the available toolbox by defining new operators and even operate on non-standard representation makes the package very flexible.

Overview of Features

• Optimization of single and multi-objective functions
• Build-in genotypes float, binary, and permutation and custom. The latter enables custom representations/genotypes
• Large collection of evolutionary operators
• Storing of the optimization process in an optimization path
• Methods to visualize results

Installation Instructions

The package will be available in a first version at CRAN soon. If you are interested in trying out and playing around with the current github developer version use the devtools package and type the following command in R:

devtools::install_github("jakobbossek/ecr")

Quickstart Example

In this section we want to optimize a one dimensional function with an evolutionary algorithm using just the evolutionary operators shipped with the package. A more in-depth introduction will be available soon.

The smoof R package provides a collection of different single objective test functions commonly used in algorithm benchmarking. As an example we are going to search for the global optimum of the one-dimensional Rastrigin function.

library(smoof)
library(ggplot2)
library(ecr)

obj.fun = makeRastriginFunction(dimensions = 1L)
autoplot(obj.fun)

As a next step we generate an ecr control object, which holds all the neccessary parameters for the evolutionary algorithm. The construction of this object needs consists of generating the object itself and kind of decorating it with some evolutionary operators.

For our setup we choose the natural representation with real-valued numbers as the genotype, a population size of 20 individuals with 5 individuals being created by recombination and mutation in each generation. Furthermore we decide to use a 'plus' survival strategy, i. e., the current population and the offspring will be merged before survival selection takes place. Gauss mutation with a standard deviance of 0.005 serves as the mutation operator and we keep the intermediate recombination operator (which is the default for representation float). Moreover we define a maximal number of 50 generations.

# Generate the control object (set basic parameters)
control = setupECRControl(
  n.population = 20L,
  n.offspring = 10L,
  representation = "float",
  survival.strategy = "plus",
  save.population.at = 0:50L,
  stopping.conditions = list(
    makeMaximumIterationsStoppingCondition(max.iter = 50L)
  )
)

# Setup the evolutionary toolbox by specifying operators
control = setupEvolutionaryOperators(
  control,
  mutator = makeGaussMutator(sdev = 0.005)
)
print(control)

Now lets start the optimization process and print the result object, which contains the optimization trace, the best parameters, the best fitness value and some additional information.

set.seed(123)
res = doTheEvolution(obj.fun, control = control)
print(res)
print(as.data.frame(res$opt.path))
print(autoplot(res, complete.trace = TRUE, log.fitness = TRUE))

Further, more complicated examples are located in the inst/examples directory.

Contact

Please address questions and missing features about the ecr package to the author Jakob Bossek [email protected]. Found some nasty bugs? Please use the issue tracker for this. Pay attention to explain the problem as good as possible. At its best you provide an example, so I can reproduce your problem.