clrng: Parallel Random Number Generation on GPU

Builds on 'gpuR' and utilizes the 'clRNG' ('OpenCL') library to provide efficient tools to generate independent random numbers in parallel on a GPU and save the results as 'R' objects, ensuring high-quality random numbers even when 'R' is used interactively or in an ad-hoc manner. Includes Fisher's simulation method adapted from Patefield, William M (1981) <doi:10.2307/2346669> and 'MRG31k3p' Random Number Generator from 'clRNG' library by Advanced Micro Devices, Inc. (2015) <https://github.com/clMathLibraries/clRNG>.

Version: 0.0.5
Depends: R (≥ 3.5.0), gpuR
LinkingTo: RViennaCL, Rcpp, gpuR, RcppEigen
Suggests: knitr
OS_type: unix
Published: 2024-07-16
Author: Ruoyong Xu ORCID iD [cre, aut, ctb], Patrick Brown ORCID iD [aut, ctb], Advanced Micro Devices, Inc. [cph] (Copyright holder of clRNG library code)
Maintainer: Ruoyong Xu <ruoyong.xu at mail.utoronto.ca>
License: GPL-3
NeedsCompilation: yes
Materials: README
CRAN checks: clrng results

Documentation:

Reference manual: clrng.pdf
Vignettes: Create streams
Fisher's exact test on GPU
Random numbers

Downloads:

Package source: clrng_0.0.5.tar.gz
Windows binaries: r-devel: not available, r-release: not available, r-oldrel: not available
macOS binaries: r-release (arm64): not available, r-oldrel (arm64): not available, r-release (x86_64): not available, r-oldrel (x86_64): not available
Old sources: clrng archive

Linking:

Please use the canonical form https://CRAN.R-project.org/package=clrng to link to this page.