Package: hmsr 1.0.1

hmsr: Multipopulation Evolutionary Strategy HMS

The HMS (Hierarchic Memetic Strategy) is a composite global optimization strategy consisting of a multi-population evolutionary strategy and some auxiliary methods. The HMS makes use of a dynamically-evolving data structure that provides an organization among the component populations. It is a tree with a fixed maximal height and variable internal node degree. Each component population is governed by a particular evolutionary engine. This package provides a simple R implementation with examples of using different genetic algorithms as the population engines. References: J. Sawicki, M. Łoś, M. Smołka, J. Alvarez-Aramberri (2022) <doi:10.1007/s11047-020-09836-w>.

Authors:Wojciech Achtelik [aut, cre], Marcin Kozubek [aut], Maciej Smołka [ths, aut], AGH University of Kraków [cph]

hmsr_1.0.1.tar.gz
hmsr_1.0.1.zip(r-4.5)hmsr_1.0.1.zip(r-4.4)hmsr_1.0.1.zip(r-4.3)
hmsr_1.0.1.tgz(r-4.4-any)hmsr_1.0.1.tgz(r-4.3-any)
hmsr_1.0.1.tar.gz(r-4.5-noble)hmsr_1.0.1.tar.gz(r-4.4-noble)
hmsr_1.0.1.tgz(r-4.4-emscripten)hmsr_1.0.1.tgz(r-4.3-emscripten)
hmsr.pdf |hmsr.html
hmsr/json (API)
NEWS

# Install 'hmsr' in R:
install.packages('hmsr', repos = c('https://wojtacht.r-universe.dev', 'https://cloud.r-project.org'))

Peer review:

Bug tracker:https://github.com/wojtacht/hms/issues

On CRAN:

3.48 score 3 stars 5 scripts 153 downloads 28 exports 29 dependencies

Last updated 9 months agofrom:f4bb133fd2. Checks:OK: 3 NOTE: 4. Indexed: yes.

TargetResultDate
Doc / VignettesOKNov 08 2024
R-4.5-winNOTENov 08 2024
R-4.5-linuxNOTENov 08 2024
R-4.4-winNOTENov 08 2024
R-4.4-macNOTENov 08 2024
R-4.3-winOKNov 08 2024
R-4.3-macOKNov 08 2024

Exports:classify_optimization_problemcma_es_metaepochdeoptim_cma_es_metaepochdeoptim_metaepochecr_metaepocheuclidean_distancega_cma_es_metaepochga_metaepochgsc_max_fitness_evaluationsgsc_metaepochs_countgsc_trivialhmslsc_max_fitness_evaluationslsc_metaepochs_without_active_childlsc_metaepochs_without_improvementlsc_trivialmanhattan_distanceplotplotActiveDemesplotPopulationprintprintBlockedSproutsprintTreertnorm_mutationsaveMetaepochsPopulationssc_max_metricshowsummary

Dependencies:cachemclicodetoolscrayonexpmfansifastmapforeachGAgenericsglueiteratorslatticelifecyclemagrittrMatrixmemoisemsmmvtnormpillarpkgconfigRcppRcppArmadillorlangsurvivaltibbleutf8uuidvctrs

Readme and manuals

Help Manual

Help pageTopics
Calculate selected ELA features for given fitness.calculate_ela_features
Classify optimization problem using selected ELA features and Random Forest model trained on BBOB dataset.classify_optimization_problem
Function that runs one cmaes metaepoch. Wrapper function for cmaes::cma_es.cma_es_metaepoch
Function that runs gradient method for one deme. Wrapper function for stats::optim.default_run_gradient_method
Function that generates run_metaepoch function for two level HMS. First level: DE, second level: CMA-ES.deoptim_cma_es_metaepoch
Function that runs one differential evolution metaepoch. Wrapper function for DEoptim::DEoptim.deoptim_metaepoch
Function that runs one ecr metaepoch. Wrapper function for ecr::ecr.ecr_metaepoch
Euclidean distanceeuclidean_distance
Function that generates run_metaepoch function for two level HMS. First level: GA, second level: CMA-ES.ga_cma_es_metaepoch
Function that runs one GA metaepoch. Wrapper function for GA::ga.ga_metaepoch
Factory function for a global stopping condition that stops the computation after fitness function has been evaluated given number of times.gsc_max_fitness_evaluations
Factory function for a global stopping condition that stops the computation after given number of metaepochs.gsc_metaepochs_count
Factory function for a global stopping condition that never stops the computation. It results in hms running until there are no more active demes.gsc_trivial
Maximization (or minimization) of a fitness function using Hierarchic Memetic Strategy.hms
A S4 class representing a result of hms.hms-class
Factory function for a local stopping condition that stops a deme after given number of fitness function evaluations has been made in that deme.lsc_max_fitness_evaluations
Factory function for a local stopping condition that stops a deme after given number of metaepochs have past since last metaepoch during which this deme had an active child.lsc_metaepochs_without_active_child
Factory function for a local stopping condition that stops a deme after given number of consecutive metaeopochs without an improvement of the best solution found in that deme.lsc_metaepochs_without_improvement
Factory function for a trivial local stopping condition that lets a deme be active forever. It is usually used in the root of a hms tree.lsc_trivial
Manhattan distancemanhattan_distance
A S4 class representing a snapshot of one metaepoch.MetaepochSnapshot-class
Plot method for "hms" class.plot,hms-method
plotActiveDemes method for "hms" class.plotActiveDemes
plotActiveDemes method for "hms" class.plotActiveDemes,hms-method
plotPopulation method for "hms" class.plotPopulation
plotPopulation method for "hms" class.plotPopulation,hms-method
Print method for class "hms".print,hms-method
printBlockedSprouts method for "hms" class.printBlockedSprouts
printBlockedSprouts method for "hms" class.printBlockedSprouts,hms-method
printTree method for class "hms".printTree
printTree method for class "hms".printTree,hms-method
Factory function that creates normal mutation functionrtnorm_mutation
saveMetaepochsPopulations method for "hms" class.saveMetaepochsPopulations
saveMetaepochsPopulationssaveMetaepochsPopulations,hms-method
Default sprouting condition based on given metric.sc_max_metric
Show method for class "hms".show,hms-method
Summary method for class "hms".summary,hms-method
Train Random Forest model on BBOB dataset (data/ela_features.rda).train_random_forest_model