Package: DiceOptim 2.1.2

DiceOptim: Kriging-Based Optimization for Computer Experiments

Efficient Global Optimization (EGO) algorithm as described in "Roustant et al. (2012)" <doi:10.18637/jss.v051.i01> and adaptations for problems with noise ("Picheny and Ginsbourger, 2012") <doi:10.1016/j.csda.2013.03.018>, parallel infill, and problems with constraints.

Authors:Mickael Binois [cre, ctb], Victor Picheny [aut], David Ginsbourger [aut], Olivier Roustant [aut], Sebastien Marmin [ctb], Tobias Wagner [ctb]

DiceOptim_2.1.2.tar.gz
DiceOptim_2.1.2.zip(r-4.7)DiceOptim_2.1.2.zip(r-4.6)DiceOptim_2.1.2.zip(r-4.5)
DiceOptim_2.1.2.tgz(r-4.6-x86_64)DiceOptim_2.1.2.tgz(r-4.6-arm64)DiceOptim_2.1.2.tgz(r-4.5-x86_64)DiceOptim_2.1.2.tgz(r-4.5-arm64)
DiceOptim_2.1.2.tar.gz(r-4.7-arm64)DiceOptim_2.1.2.tar.gz(r-4.7-x86_64)DiceOptim_2.1.2.tar.gz(r-4.6-arm64)DiceOptim_2.1.2.tar.gz(r-4.6-x86_64)
DiceOptim_2.1.2.tgz(r-4.6-emscripten)
manual.pdf |manual.html✨
card.svg |card.png
DiceOptim/json (API)

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

On CRAN:

Conda:

This package does not link to any Github/Gitlab/R-forge repository. No issue tracker or development information is available.

3.32 score 5 stars 126 scripts 3.3k downloads 42 exports 7 dependencies

Last updated from:ad73634474. Checks:13 OK. Indexed: yes.

TargetResultTimeFilesSyslog
linux-devel-arm64OK141
linux-devel-x86_64OK149
source / vignettesOK162
linux-release-arm64OK144
linux-release-x86_64OK132
macos-release-arm64OK128
macos-release-x86_64OK175
macos-oldrel-arm64OK128
macos-oldrel-x86_64OK309
windows-develOK113
windows-releaseOK124
windows-oldrelOK110
wasm-releaseOK109

Exports:AEIAEI.gradAKGAKG.gradbranin2checkPredictcrit_ALcrit_EFIcrit_SUR_cstcritcst_optimizereasyEGOeasyEGO.cstEGO.cstEGO.nstepsEIEI.gradEQIEQI.gradfastEGO.nstepsfastfungoldsteinpricehartman4integration_design_cstkriging.quantilekriging.quantile.gradmax_AEImax_AKGmax_critmax_EImax_EQImax_qEImin_quantilenoisy.optimizerParrConstraintqEGO.nstepsqEIqEI.gradrosenbrock4sampleFromEIsphere6TREGO.nstepsupdate_km_noisyEGO

Dependencies:DiceDesignDiceKrigingmnormtpbivnormrandtoolboxrgenoudrngWELL

Readme and manuals

Help Manual

Help pageTopics
Augmented Expected ImprovementAEI
AEI's GradientAEI.grad
Approximate Knowledge Gradient (AKG)AKG
AKG's GradientAKG.grad
Prevention of numerical instability for a new observationcheckPredict
Expected Augmented Lagrangian Improvementcrit_AL
Expected Feasible Improvementcrit_EFI
Stepwise Uncertainty Reduction criterioncrit_SUR_cst
Maximization of constrained Expected Improvement criteriacritcst_optimizer
User-friendly wrapper of the functions 'fastEGO.nsteps' and 'TREGO.nsteps'. Generates initial DOEs and kriging models (objects of class 'km'), and executes 'nsteps' iterations of either EGO or TREGO.easyEGO
EGO algorithm with constraintseasyEGO.cst
Sequential constrained Expected Improvement maximization and model re-estimation, with a number of iterations fixed in advance by the userEGO.cst
Sequential EI maximization and model re-estimation, with a number of iterations fixed in advance by the userEGO.nsteps
Analytical expression of the Expected Improvement criterionEI
Analytical gradient of the Expected Improvement criterionEI.grad
Expected Quantile ImprovementEQI
EQI's GradientEQI.grad
Sequential EI maximization and model re-estimation, with a number of iterations fixed in advance by the userfastEGO.nsteps
Fastfun functionfastfun
Generic function to build integration points (for the SUR criterion)integration_design_cst
Kriging quantilekriging.quantile
Analytical gradient of the Kriging quantile of level betakriging.quantile.grad
Maximizer of the Augmented Expected Improvement criterion functionmax_AEI
Maximizer of the Expected Quantile Improvement criterion functionmax_AKG
Maximization of the Expected Improvement criterionmax_crit
Maximization of the Expected Improvement criterionmax_EI
Maximizer of the Expected Quantile Improvement criterion functionmax_EQI
Maximization of multipoint expected improvement criterion (qEI)max_qEI
Minimization of the Kriging quantile.min_quantile
Optimization of homogenously noisy functions based on Krigingnoisy.optimizer
2D constraint functionParrConstraint
Sequential multipoint Expected improvement (qEI) maximizations and model re-estimationqEGO.nsteps
Analytical expression of the multipoint expected improvement (qEI) criterionqEI
Gradient of the multipoint expected improvement (qEI) criterionqEI.grad
Sampling points according to the expected improvement criterionsampleFromEI
Test constraints violation (vectorized)test_feas_vec
Trust-region based EGO algorithm.TREGO.nsteps
Update of one or two Kriging models when adding new observationupdate_km_noisyEGO