Package: hetGP 1.1.9

hetGP: Heteroskedastic Gaussian Process Modeling and Design under Replication

Performs Gaussian process regression with heteroskedastic noise following the model by Binois, M., Gramacy, R., Ludkovski, M. (2016) <doi:10.48550/arXiv.1611.05902>, with implementation details in Binois, M. & Gramacy, R. B. (2021) <doi:10.18637/jss.v098.i13>. The input dependent noise is modeled as another Gaussian process. Replicated observations are encouraged as they yield computational savings. Sequential design procedures based on the integrated mean square prediction error and lookahead heuristics are provided, and notably fast update functions when adding new observations.

Authors:Mickael Binois [aut, cre], Robert B. Gramacy [aut]

hetGP_1.1.9.tar.gz
hetGP_1.1.9.zip(r-4.7)hetGP_1.1.9.zip(r-4.6)hetGP_1.1.9.zip(r-4.5)
hetGP_1.1.9.tgz(r-4.6-x86_64)hetGP_1.1.9.tgz(r-4.6-arm64)hetGP_1.1.9.tgz(r-4.5-x86_64)hetGP_1.1.9.tgz(r-4.5-arm64)
hetGP_1.1.9.tar.gz(r-4.7-arm64)hetGP_1.1.9.tar.gz(r-4.7-x86_64)hetGP_1.1.9.tar.gz(r-4.6-arm64)hetGP_1.1.9.tar.gz(r-4.6-x86_64)
hetGP_1.1.9.tgz(r-4.6-emscripten)
manual.pdf |manual.html✨
DESCRIPTION |NEWS
card.svg |card.png
hetGP/json (API)

# Install 'hetGP' in R:

install.packages('hetGP', repos = c('https://mbinois.r-universe.dev', 'https://cloud.r-project.org'))

Uses libs:

c++– GNU Standard C++ Library v3

Datasets:

ato.a - Assemble To Order (ATO) Data and Fits
bfs.exp - Bayes Factor Data
bfs.gamma - Bayes Factor Data
kill - Assemble To Order (ATO) Data and Fits
mult - Assemble To Order (ATO) Data and Fits
nc - Assemble To Order (ATO) Data and Fits
out - Assemble To Order (ATO) Data and Fits
out.a - Assemble To Order (ATO) Data and Fits
train - Assemble To Order (ATO) Data and Fits
X - Assemble To Order (ATO) Data and Fits
Xa - Assemble To Order (ATO) Data and Fits
Xtest - Assemble To Order (ATO) Data and Fits
Xtrain - Assemble To Order (ATO) Data and Fits
Xtrain.out - Assemble To Order (ATO) Data and Fits
Z - Assemble To Order (ATO) Data and Fits
Za - Assemble To Order (ATO) Data and Fits
Zm - Assemble To Order (ATO) Data and Fits
Ztest - Assemble To Order (ATO) Data and Fits
Ztrain - Assemble To Order (ATO) Data and Fits
Ztrain.out - Assemble To Order (ATO) Data and Fits
Zv - Assemble To Order (ATO) Data and Fits

On CRAN:

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

cpp

5.11 score 5 stars 3 packages 288 scripts 838 downloads 46 exports 5 dependencies

Last updated from:6df210a8bd. Checks:13 OK. Indexed: yes.

Target	Result	Time
linux-devel-arm64	OK	161
linux-devel-x86_64	OK	178
source / vignettes	OK	393
linux-release-arm64	OK	165
linux-release-x86_64	OK	170
macos-release-arm64	OK	136
macos-release-x86_64	OK	285
macos-oldrel-arm64	OK	140
macos-oldrel-x86_64	OK	285
windows-devel	OK	203
windows-release	OK	225
windows-oldrel	OK	174
wasm-release	OK	112

Exports:allocate_mult allocq allocq_c compareGP cov_gen crit_cSUR crit_EI crit_ICU crit_IMSPE crit_logEI crit_MCU crit_MEE crit_optim crit_qEI crit_tMSE deriv_crit_EI deriv_crit_IMSPE deriv_crit_logEI f1d f1d_n f1d2 f1d2_n find_reps horizon hyperSharpeMax hyperSharperQ IMSPE IMSPE_optim logLikH LOO_preds mleCRNGP mleHetGP mleHetTP mleHomGP mleHomTP pred_noisy_input qEI_loop qHSRI_loop rebuild scores simul sirEval sirSimulate strip update_pred Wij

Dependencies:DiceDesign MASS mco quadprog Rcpp

a guide to the hetGP package

Rendered fromhetGP_vignette.Rnwusingknitr::knitr

Last update: 2026-04-23
Started: 2019-01-10

Help page	Topics
Allocation of replicates on existing designs	allocate_mult
Allocation	allocq
Allocation under maximum replication constraints	allocq_c
Assemble To Order (ATO) Data and Fits	ato ato.a kill mult nc out out.a reps train X Xa Xtest Xtrain Xtrain.out Z Za Zm Ztest Ztrain Ztrain.out Zv
Bayes Factor Data	bfs bfs.exp bfs.gamma
Likelihood-based comparison of models	compareGP
Correlation function of selected type, supporting both isotropic and product forms	cov_gen
Contour Stepwise Uncertainty Reduction criterion	crit_cSUR
Expected Improvement criterion	crit_EI
Integrated Contour Uncertainty criterion	crit_ICU
Sequential IMSPE criterion	crit_IMSPE
Logarithm of Expected Improvement criterion	crit_logEI
Maximum Contour Uncertainty criterion	crit_MCU
Maximum Empirical Error criterion	crit_MEE
Criterion optimization	crit_optim
Parallel Expected improvement	crit_qEI
t-MSE criterion	crit_tMSE
Derivative of EI criterion for GP models	deriv_crit_EI
Derivative of crit_IMSPE	deriv_crit_IMSPE
Derivative of logEI criterion for GP models	deriv_crit_logEI
1d test function (1)	f1d
Noisy 1d test function (1) Add Gaussian noise with variance r(x) = scale * (1.1 + sin(2 pi x))^2 to 'f1d'	f1d_n
1d test function (2)	f1d2
Noisy 1d test function (2) Add Gaussian noise with variance r(x) = scale * (exp(sin(2 pi x)))^2 to 'f1d2'	f1d2_n
Data preprocessing	find_reps
Adapt horizon	horizon
Hypervolume Sharpe ratio maximization	hyperSharpeMax
Hypervolume Sharpe ratio return and covariance	hyperSharperQ
Integrated Mean Square Prediction Error	IMSPE
IMSPE optimization	IMSPE_optim
Leave one out predictions	LOO_preds
Gaussian process modeling with correlated noise	mleCRNGP
Gaussian process modeling with heteroskedastic noise	mleHetGP
Student-t process modeling with heteroskedastic noise	mleHetTP
Gaussian process modeling with homoskedastic noise	mleHomGP
Student-T process modeling with homoskedastic noise	mleHomTP
Gaussian process prediction prediction at a noisy input 'x', with centered Gaussian noise of variance 'sigma_x'. Several options are available, with different efficiency/accuracy tradeoffs.	pred_noisy_input
Gaussian process predictions using a GP object for correlated noise (of class 'CRNGP')	predict.CRNGP
Gaussian process predictions using a heterogeneous noise GP object (of class 'hetGP')	predict.hetGP
Student-t process predictions using a heterogeneous noise TP object (of class 'hetTP')	predict.hetTP
Gaussian process predictions using a homoskedastic noise GP object (of class 'homGP')	predict.homGP
Student-t process predictions using a homoskedastic noise GP object (of class 'homGP')	predict.homTP
BO loop with qEI	qEI_loop
BO loop with massive batches	qHSRI_loop
Import and export of hetGP objects	rebuild rebuild.hetGP rebuild.hetTP rebuild.homGP rebuild.homTP strip
Score and RMSE function To asses the performance of the prediction, this function computes the root mean squared error and proper score function (also known as negative log-probability density).	scores
Conditional simulation for CRNGP	simul
Fast conditional simulation for a CRNGP model	simul.CRNGP
SIR test problem	sirEval sirSimulate
Prediction update with new designs and observations	update_pred
Update '"hetGP"'-class model fit with new observations	update.hetGP
Update '"hetTP"'-class model fit with new observations	update.hetTP
Fast 'homGP'-update	update.homGP
Fast 'homTP'-update	update.homTP
Compute double integral of the covariance kernel over a [0,1]^d domain	Wij

Package: hetGP 1.1.9

hetGP: Heteroskedastic Gaussian Process Modeling and Design under Replication

Citation

Readme and manuals

Help Manual

Usage by other packages (reverse dependencies)