Interface to 'libKriging' 'C++' library < https://github.com/libKriging> that should provide most standard Kriging / Gaussian process regression features (like in 'DiceKriging', 'kergp' or 'RobustGaSP' packages). 'libKriging' relies on Armadillo linear algebra library (Apache 2 license) by Conrad Sanderson, 'lbfgsb_cpp' is a 'C++' port around by Pascal Have of 'lbfgsb' library (BSD-3 license) by Ciyou Zhu, Richard Byrd, Jorge Nocedal and Jose Luis Morales used for hyperparameters optimization.
This repository is just a wrapper to emulate a "standard" R package from https://github.com/libKriging/bindings/R/rlibkriging content, so you can install it just using:
install.packages('devtools')
devtools::install_github("libKriging/rlibkriging")
# Or install a specific tagged version
devtools::install_github("libKriging/[email protected]")
Note: When installing via install_github(), the package automatically initializes git submodules (src/libK, src/slapack) at the correct versions using the pinned commits recorded in tools/gitmodules-shas. This ensures you get the exact versions of dependencies that were used when the release was tagged.
The stable version is available from CRAN :
install.packages('rlibkriging')
c++ and cmake, should be installed using:
brew/apt/yum/... install cpp cmakeNote: this repository mainly contains modified Makefiles, inspired by https://github.com/astamm/nloptr wrapper.
When submitting to CRAN, ./tools/setup.sh should be run before R CMD build rlibkriging to fit CRAN policy.
For maintainers: when updating submodules, always run ./tools/update_submodule_shas.sh to record the new commit SHAs. This ensures users who install via install_github() get the correct submodule versions. See SUBMODULE_VERSION_MANAGEMENT.md for details.