Source: r-cran-loo
Standards-Version: 4.7.4
Maintainer: Debian R Packages Maintainers <r-pkg-team@alioth-lists.debian.net>
Uploaders:
 Andreas Tille <tille@debian.org>,
Section: gnu-r
Testsuite: autopkgtest-pkg-r
Build-Depends:
 debhelper-compat (= 14),
 dh-r,
 r-base-dev,
 r-cran-checkmate,
 r-cran-matrixstats,
 r-cran-posterior (>= 1.7.0),
 architecture-is-64-bit,
 architecture-is-little-endian,
Vcs-Browser: https://salsa.debian.org/r-pkg-team/r-cran-loo
Vcs-Git: https://salsa.debian.org/r-pkg-team/r-cran-loo.git
Homepage: https://cran.r-project.org/package=loo

Package: r-cran-loo
Architecture: all
Depends:
 ${R:Depends},
 ${misc:Depends},
Recommends:
 ${R:Recommends},
Suggests:
 ${R:Suggests},
Description: Efficient Leave-One-Out Cross-Validation and WAIC for Bayesian Models
 Efficient approximate leave-one-out cross-validation (LOO) for Bayesian
 models fit using Markov chain Monte Carlo, as described in Vehtari, Gelman,
 and Gabry (2017) <doi:10.1007/s11222-016-9696-4>. The approximation uses
 Pareto smoothed importance sampling (PSIS), a new procedure for regularizing
 importance weights. As a byproduct of the calculations, we also obtain
 approximate standard errors for estimated predictive errors and for the
 comparison of predictive errors between models. The package also provides
 methods for using stacking and other model weighting techniques to average
 Bayesian predictive distributions.
