laGP: Local Approximate Gaussian Process Regression

Performs approximate GP regression for large computer experiments and spatial datasets. The approximation is based on finding small local designs for prediction (independently) at particular inputs. OpenMP and SNOW parallelization are supported for prediction over a vast out-of-sample testing set; GPU acceleration is also supported for an important subroutine. OpenMP and GPU features may require special compilation. An interface to lower-level (full) GP inference and prediction is also provided, as are associated wrapper routines for blackbox optimization under constraints via an augmented Lagrangian scheme, and large scale computer model calibration.

Version: 1.1-5
Depends: R (≥ 2.14)
Imports: tgp, parallel
Suggests: mvtnorm, MASS, akima, lhs, crs
Published: 2015-05-27
Author: Robert B. Gramacy
Maintainer: Robert B. Gramacy <rbgramacy at>
License: LGPL-2 | LGPL-2.1 | LGPL-3 [expanded from: LGPL]
NeedsCompilation: yes
Materials: README ChangeLog INSTALL
CRAN checks: laGP results


Reference manual: laGP.pdf
Vignettes: a guide to the laGP package
Package source: laGP_1.1-5.tar.gz
Windows binaries: r-devel:, r-release:, r-oldrel:
OS X Snow Leopard binaries: r-release: laGP_1.1-5.tgz, r-oldrel: laGP_1.1-3.tgz
OS X Mavericks binaries: r-release: laGP_1.1-5.tgz
Old sources: laGP archive

Reverse dependencies:

Reverse suggests: mlr