Package: STPGA 7.0.2

STPGA: Selection of Training Populations by Genetic Algorithm

Advanced genetic algorithms for optimal subset selection in high-dimensional prediction problems. Provides efficient single and multi-objective optimization for training population selection with comprehensive criteria including A-, D-, E-optimality, prediction error variance (PEV), Cook's distance (CD), and distance-based measures. Features multi-criteria convergence detection, restart mechanisms, rank-based selection with pressure control, adaptive mutation, diversity preservation, and numerically stable matrix operations. Includes convergence diagnostics, configurable optimization windows, and both modern clean interfaces with legacy compatibility functions.

Authors:Deniz Akdemir

STPGA_7.0.2.tar.gz
STPGA_7.0.2.zip(r-4.7)STPGA_7.0.2.zip(r-4.6)STPGA_7.0.2.zip(r-4.5)
STPGA_7.0.2.tgz(r-4.6-any)STPGA_7.0.2.tgz(r-4.5-any)
STPGA_7.0.2.tar.gz(r-4.7-any)STPGA_7.0.2.tar.gz(r-4.6-any)
STPGA_7.0.2.tgz(r-4.6-emscripten)
manual.pdf |manual.html✨
card.svg |card.png
STPGA/json (API)
NEWS

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

Bug tracker:https://github.com/denizakdemir/stpga/issues

Datasets:
  • Wheat.K - Adult plant height (estimated genetic values) for 1182 elite wheat lines
  • Wheat.M - Adult plant height (estimated genetic values) for 1182 elite wheat lines
  • Wheat.Y - Adult plant height (estimated genetic values) for 1182 elite wheat lines

On CRAN:

Conda:

5.06 score 36 scripts 216 downloads 8 mentions 90 exports 13 dependencies

Last updated from:d307a34778. Checks:7 ERROR, 2 OK. Indexed: yes.

TargetResultTimeFilesSyslog
linux-devel-x86_64ERROR157
source / vignettesOK283
linux-release-x86_64ERROR152
macos-release-arm64ERROR116
macos-oldrel-arm64ERROR162
windows-develERROR112
windows-releaseERROR109
windows-oldrelERROR107
wasm-releaseOK110

Exports:a_optimalityadaptive_mutation_rateAmat.piecesblock_matrix_multcache_statscalculate_crowding_distancecalculate_diversitycalculate_population_distancescd_meancd_mean_mmcd_mean_mm_h2check_matrix_dimensionscheck_memory_usagecheck_multi_criteria_convergencecheck_parameter_namingchunked_crossprodchunked_matrix_operationclear_cachecompute_amatrixcompute_population_statscompute_prediction_coreconvergence_diagnosticscreate_function_aliascreate_progress_barcriterioncrossovercrowding_replacementd_optimalitydist_to_testdist_to_test2distance_criteriondistance_internal_meandistance_internal_mindistance_to_idealdistance_train_to_test_maxdistance_train_to_test_meandisttoidealdiversity_summarydocument_parametere_optimalityelite_selectionevaluate_populationevaluate_population_legacyevaluate_population_optimizedevaluate_population_smartfind_non_dominatedfinish_progressfitness_sharingg_optimalityGenAlgForSubsetSelectionMOGenAlgForSubsetSelectionMONoTestgenerate_doc_templategenerate_offspringgenomic_relationship_matrixget_adaptive_ridgeget_function_variantsget_optimal_coresh2_to_variancesi_optimalityinfluence_measure_legacymatrix_stability_checkmutate_solutionneg_dist_in_trainneg_dist_in_train2non_dominated_sortingparallel_applyperform_restartpev_maxpev_meanpev_mean_mmpev_mean_mm_h2population_distancesrank_selectionridge_regression_cvsafe_matrix_inversesimple_cachestandardize_namestandardize_parameterssubset_gasubset_ga_multiobjectivesubset_ga_multiobjective_singlesubset_ga_singletournament_selectiontransform_fitnessunified_distance_criterionupdate_progressvalidate_kinship_matrixvalidate_matrix_paramsvalidate_selection_parametersvalidate_variance_components

Dependencies:AlgDesigncliemoafarvergluelabelinglifecycleR6RColorBrewerrlangscalesscatterplot3dviridisLite

STPGA Package Demo Report

Deniz Akdemir

Rendered fromsimple_demo.Rmdusingknitr::rmarkdownon May 28 2026.

Last update: 2026-05-28
Started: 2025-07-27

STPGA: Subset Training Population Genetic Algorithm

Deniz Akdemir

Rendered fromSTPGA-package.Rmdusingknitr::rmarkdownon May 28 2026.

Last update: 2026-05-28
Started: 2025-07-27

Readme and manuals

Help Manual

Help pageTopics
A-optimality criterion for experimental designa_optimality
Adaptive mutation rate based on population diversityadaptive_mutation_rate
Efficient block matrix multiplicationblock_matrix_mult
Get cache statisticscache_stats
Calculate crowding distance for diversity preservationcalculate_crowding_distance
Calculate population diversitycalculate_diversity
Coefficient of Determination (R²) based on experimental design literaturecd_mean
Coefficient of Determination (R²) for Mixed Modelscd_mean_mm
Coefficient of Determination for Mixed Models using heritabilitycd_mean_mm_h2
Check convergence based on convergence historycheck_convergence
Check dominance between two solutionscheck_dominance
Check matrix dimensions for operationscheck_matrix_dimensions
Check memory usage and warn for large computationscheck_memory_usage
Multi-criteria convergence detectioncheck_multi_criteria_convergence
Check if parameter names follow conventionscheck_parameter_naming
Memory-efficient crossprod computationchunked_crossprod
Compute matrix operations in chunks for memory efficiencychunked_matrix_operation
Clear computation cacheclear_cache
STPGA Matrix Operationscompute_amatrix
Compute convergence metrics for multi-objective optimizationcompute_convergence_metrics
Compute generation statistics for multi-objective optimizationcompute_generation_stats
Population statistics computationcompute_population_stats
Compute prediction core for matrix operationscompute_prediction_core
Get convergence history diagnostics from GA resultsconvergence_diagnostics
Create evaluation cache for storing computed fitness valuescreate_evaluation_cache
Create function alias for backward compatibilitycreate_function_alias
Performance Optimization Functionscreate_progress_bar
Unified criterion function for optimizationcriterion
STPGA Genetic Algorithm Operatorscrossover
Crowding replacement for diversity preservationcrowding_replacement
D-optimality criterion for experimental designd_optimality
Determine optimal batch size for evaluationdetermine_optimal_batch_size
Determine primary matrix for optimizationdetermine_primary_matrix
Unified distance criterion functiondistance_criterion
Negative mean distance within training setdistance_internal_mean
Negative minimum distance within training setdistance_internal_min
STPGA Distance Functionsdistance_to_ideal
Maximum distance from training to test setdistance_train_to_test_max
Mean distance from training to test setdistance_train_to_test_mean
Diversity-preserving selection for environmental selectiondiversity_preserving_selection
Population diversity summarydiversity_summary
Generate consistent parameter documentationdocument_parameter
E-optimality criterion for experimental designe_optimality
Elite selectionelite_selection
Environmental selection for NSGA-IIenvironmental_selection
Evaluate population for multiple objectivesevaluate_multiobjective_population
STPGA Population Evaluation Functionsevaluate_population
Smart population evaluation with adaptive batch processingevaluate_population_smart
Evaluate a batch of solutions efficientlyevaluate_solution_batch
Extract final Pareto frontextract_pareto_front
Find non-dominated solutionsfind_non_dominated
Finish progress barfinish_progress
Fitness sharing for diversity preservationfitness_sharing
G-optimality criterion (minimizes maximum prediction variance)g_optimality
Generate standard function documentation templategenerate_doc_template
Generate offspring for multi-objective optimizationgenerate_multiobjective_offspring
Generate offspring from elite solutionsgenerate_offspring
Compute genomic relationship matrix using different methodsgenomic_relationship_matrix
Get adaptive ridge parameter based on condition numberget_adaptive_ridge
Get consistent function name variantsget_function_variants
Get optimal number of cores for computationget_optimal_cores
Group similar solutions for batch processinggroup_similar_solutions
STPGA Optimization Criteria Functionsh2_to_variances
I-optimality criterion (minimizes average prediction variance)i_optimality
Legacy influence measure (for backward compatibility)influence_measure_legacy
Initialize random populationinitialize_population
Legacy compatibility wrappersAmat.pieces calculate_population_distances disttoideal dist_to_test dist_to_test2 evaluate_population_legacy evaluate_population_optimized GenAlgForSubsetSelectionMO GenAlgForSubsetSelectionMONoTest legacy-compatibility neg_dist_in_train neg_dist_in_train2 unified_distance_criterion
Compute condition number and numerical stability metricsmatrix_stability_check
Multi-objective selection based on dominance ranksmultiobjective_selection
Mutate a solutionmutate_solution
Non-dominated sorting for multi-objective optimizationnon_dominated_sorting
Parallel matrix operations with proper error handlingparallel_apply
Perform genetic algorithm restart with population diversificationperform_restart
Maximum Prediction Error Variance (PEV)pev_max
Mean Prediction Error Variance (PEV) - Literature Correctedpev_mean
Mean Prediction Error Variance for Mixed Models (Henderson's BLUP)pev_mean_mm
Mean Prediction Error Variance for Mixed Models using heritabilitypev_mean_mm_h2
Plot Pareto front progressplot_pareto_progress
Calculate population diversity metricspopulation_distances
Rank-based selection with selection pressure controlrank_selection
Ridge regression with optimal lambda selectionridge_regression_cv
Efficient matrix inversion with numerical stabilitysafe_matrix_inverse
Select diverse subset using crowding distanceselect_diverse_subset
Simple hash-based cache for computation resultssimple_cache
Naming Standards and Conventionsstandardize_name
Standardize common parameter names across functionsstandardize_parameters
STPGA Core Genetic Algorithm Functionssubset_ga
STPGA Multi-Objective Genetic Algorithmsubset_ga_multiobjective
Multi-objective GA without test set (single population)subset_ga_multiobjective_single
Genetic algorithm for subset selection without test set (single objective)subset_ga_single
Tournament selectiontournament_selection
Fitness transformation and normalization utilitiestransform_fitness
Update Pareto archive with new solutionsupdate_pareto_archive
Update progress barupdate_progress
Validate kinship matrixvalidate_kinship_matrix
Input Validation and Utility Functionsvalidate_matrix_params
Validate and adjust selection parametersvalidate_selection_parameters
Validate variance componentsvalidate_variance_components
Adult plant height (estimated genetic values) for 1182 elite wheat linesWheat.K Wheat.M Wheat.Y WheatData