rjuliabugs is an R package that provides a bridge between R and JuliaBUGS, the BUGS-style Bayesian modeling interface developed in Julia as part of the Turing.jl probabilistic programming ecosystem.
JuliaBUGS allows users to define models using the familiar BUGS syntax while leveraging the speed and flexibility of the Julia language. It enables automatic translation of BUGS code into modern probabilistic programs that run efficiently with advanced inference engines such as Hamiltonian Monte Carlo (HMC) — all accessible via Turing.jl.
With rjuliabugs, R users can run BUGS models/code
through JuliaBUGS and take advantage of the inference algorithms
available in Turing.jl, without leaving the R environment. This not only
provides a seamless path to adopt faster (including parallelization)
and more flexible sampling methods but also allows integration with R’s
extensive post-processing ecosystem (e.g., bayesplot,
posterior, coda). It lowers the barrier for
existing BUGS users to adopt modern Bayesian tools without abandoning
their existing model codebase.
You can install the development version of rjuliabugs from GitHub with:
# install.packages("remotes")
remotes::install_github("MateusMaiaDS/rjuliabugs")⚠️ Loading the package:
When starting a R session for the first time, after loading the package
rjuliabugs and before running any function, we recommend to
run rjuliabugs::setup_juliaBUGS() to build all Julia setup
in R and install any needed dependencies. We recommend this initially as
it an be time-consuming. See ?rjuliabugs::setup_juliaBUGS
for more details.
For further guidance, there articles in this page which summarize
most of the FAQ and additional instructions for installing Julia and
setting rjuliabugs, troubleshooting, and running
rjuliabugs in parallel. We recommend the reading of them
before working with rjuliabugs:
Contributions are welcome! If you encounter a bug or have a feature request, please open an issue.
This package is licensed under the MIT License. See the LICENSE file for details.
This project is part of a Google Summer of Code 2025 initiative under the Turing.jl organization. The mentors were Xianda Sun and Robert Goudie.