Changes in version 3.2.5 (2026-01-08)                  

  - Fixed bug in C++ code for optimProbsel_Rcpp() by rewriting
    rank_asR().

                 Changes in version 3.2.4 (2024-01-28)                  

  - Fixed CRAN issues.

                 Changes in version 3.2.3 (2022-10-18)                  

  - Fixed warnings in C++ code under CRAN request.

                 Changes in version 3.2.2 (2021-10-15)                  

  - Fixed doi in JSS and RJ publication.

                 Changes in version 3.2.1 (2021-04-21)                  

  - Forced elitism argument to be an integer.
  - Fixed passing further fitness function arguments in de().
  - Fixed passing suggestions in gaisl().
  - Added an example of postFitness usage in the vignette.
  - Added rmarkdown in the list of suggested package (as required by
    CRAN).

                  Changes in version 3.2 (2019-01-10)                   

  - Added website using pkgdown.
  - Added a section to vignette on integer optimisation.
  - Added function de() implementing Differential Evolution based on the
    description in Simon (2013) Evolutionary Optimization Algorithms.
  - Bug fix in C++ code calling pow() function.
  - Bug fix in gails() when suggestions are provided.
  - Long outputs in summary() function calls are shortened.
  - Fix a bug in gaperm_pbxCrossover().

                 Changes in version 3.1.1 (2018-05-11)                  

  - Bug fix in C++ code calling pow() function.

                  Changes in version 3.1 (2018-05-09)                   

  - Genetic operators available in C++ using Rcpp package.
  - Add parameter "useRcpp" in gaControl() to control if the C++
    implementation of genetic operators should be used. By default is
    set to TRUE.
  - Renamed input parameters min and max to lower and upper. The old
    nomenclature is still accepted but deprecated.
  - Added stopParallel() function to stop a cluster if parallel = TRUE,
    including registerDoSEQ() to avoid problems if foreach loop is used
    after.
  - A parallel cluster is automatically stopped unless a registered
    parallel back end is provided as argument to parallel argument in
    the ga() or gaisl() function call.
  - Bug fix in gaperm_Population() when min > 1. Thanks to Romero
    Barata.
  - Update "A quick tour of GA" vignette.
  - Add README.Rmd that is shown in GitHub and CRAN.
  - Create a sticker for the GA package.

                 Changes in version 3.0.2 (2016-06-07)                  

  - In interactive sessions use different monitor functions depending on
    whether or not is an RStudio session. This is due to different
    capabilities of the several consoles from which R can be run.

                 Changes in version 3.0.1 (2016-05-16)                  

  - doRNG package moved from Depends to Suggests due to the
    unavailability of the binary version of the package on Windows. As a
    consequence, GA package can be installed even without doRNG being
    installed, but in this case results are not reproducible by setting
    the seed if GAs are executed in parallel.

                  Changes in version 3.0 (2016-05-10)                   

  - Added option to provide a cluster for parallelisation.
  - Added GA-hybrid using optim() for local search.
  - Added GA Islands model in gaisl().
  - Rewrite of gaMonitor() to clear the previous output before printing
    the info about the current iteration. Old version is available in
    gaMonitor2(). Same behaviour for gaislMonitor() and gaislMonitor2().

                         Changes in version 2.3                         

  - Added documentation vignettes.

                  Changes in version 2.2 (2014-10-15)                   

  - Slight modification to plot.ga() method.
  - Included link in main ga documentation to genetic operators.
  - Added ga_pmutation() function to allow GAs having variable mutation
    probability.
  - pdf files (previously included as vignettes) moved to inst/doc with
    corresponding index.html.

                  Changes in version 2.1 (2014-05-06)                   

  - .printShortMatrix() is a function to print part of rows/columns of a
    matrix
  - print.summary.ga() accept arguments to be passed to
    .printShortMatrix(). This allow to shorten the printed output in
    case of large dimensions for the matrices containing the suggestions
    and the final solutions.
  - plot.ga() now shadowed the area between the max and the median of
    fitness values at each iteration. Changed to pch from 17 to 1 for
    means.
  - Function gaSummary() is embedded in the code, so it cannot be
    defined by the user.
  - Added function ga_pmutation() for computing variable mutation
    probability.
  - Modified main function ga() to allow for pmutation to be a function.
    This enables the use of variable mutation rate.
  - Export plot.ga class in NAMESPACE and used as S4 method.
  - Computing summaries at each step is done by function gaSummary(),
    but it may be defined by the user.
  - Export summary.ga class in NAMESPACE and used as S4 method.
  - Default for suggestions argument in ga() is set to NULL.
  - Fixed a bug in gabin_uCrossover().
  - Added description of slots in the help page for ga-class.
  - Bug fix on setting maxfitness.
  - Bug fix on passing seed argument when "snow" parallel is used.
  - Add explicitly stop clusters if parallel is used.
  - Parallel computing improving set up of clusters.

                  Changes in version 2.0 (2013-08-20)                   

  - Option for parallel computing.
  - Added argument keepBest to include the best solution at each
    iteration.
  - Bug fix when using suggestions argument.

                  Changes in version 1.1 (2013-04-08)                   

  - Update citation and references to JSS paper.

                  Changes in version 1.0 (2012-06-22)                   

  - First release on CRAN.