Repository files navigation

⚠ Important Bug Fix for CRAN Release 2.6.0

If you encounter Python import errors related tolibstdc++.so.6 on Linux orlibcblas.so.3 on Intel-based machines with CRAN release 2.6.0, the issue isfixed in the development version (2.6.0-9000).Please install the development version following theinstallation instructions below.The next CRAN release will include this fix.

The R packagedgpsi provides R interface to Python packagedgpsi for deep and linked Gaussian process emulations using stochastic imputation (SI).

Hassle-free Python Setup
You don't need prior knowledge of Python to start using the package, all you need is a single click in R (seeInstallation section below) that automatically installs and activates the required Python environment for you!

dgpsi currently has following features:

• Gaussian process emulations with separable or non-separable squared exponential and Matérn-2.5 kernels.
• Deep Gaussian process emulations with flexible structures including:
  • multiple layers;
  • multiple GP nodes;
  • separable or non-separable squared exponential and Matérn-2.5 kernels;
  • global input connections;
  • non-Gaussian likelihoods (Poisson, Negative-Binomial, heteroskedastic Gaussian, and Categorical).
• Linked emulations of feed-forward systems of computer models by linking (D)GP emulators of deterministic individual computer models.
• Fast Leave-One-Out (LOO) and Out-Of-Sample (OOS) validations for GP, DGP, and linked (D)GP emulators.
• Multi-core predictions and validations for GP, DGP, and Linked (D)GP emulators.
• Sequential designs for (D)GP emulators and bundles of (D)GP emulators.
• Automatic pruning of DGP emulators, both statically and dynamically.
• Large-scale GP, DGP, and Linked (D)GP emulations.
• Scalable DGP classification using Stochastic Imputation.
• Bayesian optimization.

• CheckA Quick Guide to dgpsi to get started with the package.
• For experimental features, check out ourwebsite for the development version.

You can install the package from CRAN:

install.packages('dgpsi')

or its development version from GitHub:

devtools::install_github('mingdeyu/dgpsi-R')

After the installation, run

library(dgpsi)

to load the package. To install or activate the required Python environment automatically, you can either rundgpsi::init_py() explicitly or simply call any function from the package. That's it - the package is ready to use!

Note
After loadingdgpsi, the package may take some time to compile and initiate the underlying Python environment the firsttime a function fromdgpsi is executed. Any subsequent function calls won't require re-compiling or re-activation of thePython environment, and will be faster.

If you experience Python related issues while using the package, please try to reinstall the Python environment:

dgpsi::init_py(reinstall=T)

Or uninstall completely the Python environment:

dgpsi::init_py(uninstall=T)

and then reinstall:

dgpsi::init_py()

This package is part of an ongoing research initiative. For detailed information about the research aspects and guidelines for use, please refer to ourResearch Notice.

Ming, D. and Williamson, D. (2023) Linked deep Gaussian process emulation for model networks. arXiv:2306.01212

Ming, D., Williamson, D., and Guillas, S. (2023) Deep Gaussian process emulation using stochastic imputation.Technometrics. 65(2), 150-161.

Ming, D. and Guillas, S. (2021) Linked Gaussian process emulation for systems of computer models using Matérn kernels and adaptive design,SIAM/ASA Journal on Uncertainty Quantification. 9(4), 1615-1642.