Package: gpumetropolis Title: GPU-Portable Vendor-Agnostic Metropolis-Hastings Sampler Version: 0.5.3 Authors@R: person("Pedro", "Carvalho Brom", , "pcbrom@gmail.com", role = c("aut", "cre", "cph"), comment = c(ORCID = "0000-0002-1288-7695")) Description: A generic random-walk Metropolis-Hastings sampler for Markov chain Monte Carlo. The user declares a model by writing its log-likelihood and log-prior as one-sided formulas; the package compiles them to a stack-machine bytecode that a single portable kernel interprets, so the same model runs on the CPU and on the GPU from one source. The sampler advances many independent chains in one batched pass. A distributional equivalence harness based on the two-sample Kolmogorov-Smirnov test and the split R-hat statistic of Gelman and Rubin (1992) checks the GPU output against the CPU output. Dispatching the kernel to a GPU helps when many chains are run or when the log-density is expensive to evaluate over a large data set, and does not help for small models run with few chains. License: MIT + file LICENSE Encoding: UTF-8 Roxygen: list(markdown = TRUE) RoxygenNote: 8.0.0 Config/rextendr/version: 0.5.0 SystemRequirements: Cargo (Rust's package manager), rustc >= 1.85.0, xz Depends: R (>= 4.2) Suggests: testthat (>= 3.0.0), knitr, rmarkdown, evd, diptest, loo, coda, MCMCpack, mcmc, nimble, BayesianTools Config/testthat/edition: 3 VignetteBuilder: knitr URL: https://github.com/pcbrom/gpumetropolis BugReports: https://github.com/pcbrom/gpumetropolis/issues Config/pak/sysreqs: xz-utils libclang-dev Repository: https://pcbrom.r-universe.dev Date/Publication: 2026-07-16 17:25:39 UTC RemoteUrl: https://github.com/pcbrom/gpumetropolis RemoteRef: HEAD RemoteSha: 2ccfe2873b945ae6b1c55a04eebee8cd0f398d89 NeedsCompilation: yes Packaged: 2026-07-16 18:55:57 UTC; root Author: Pedro Carvalho Brom [aut, cre, cph] (ORCID: ) Maintainer: Pedro Carvalho Brom