A R package that wraps the LPJ-GUESS model
Install the latest stable release from https://github.com/biometry/rLPJGUESS/releases.
You can download the binary and install it as follows
install.packages("/path/to/binary/rLPJGUESS_1.1.0.tar.gz", repos = NULL, type = "source")
Or you can install it directly from the download link
library(devtools)
install_url("https://github.com/biometry/rLPJGUESS/releases/download/v1.1.0/rLPJGUESS_1.1.0.tar.gz",
dependencies = T, build_vignettes = T)
library(rLPJGUESS)
?rLPJGUESS
Note: rLPJGUESS relies on two packages for parallelization:
- snow (for SOCK cluster, if you use PC/laptop)
- Rmpi (for MPI clusters)
Installing "Rmpi" might be complicated, and it is not strictly required: If you are thinking of using rLPJGUESS on a laptop or workstation, you will be dealing with SOCK clusters and you do not need Rmpi.
If you want to install the current (development) version from this repository, run
devtools::install_github(repo = "biometry/rLPJGUESS", subdir = "rLPJGUESS",
dependencies = T, build_vignettes = T)
Below the status of the automatic Travis CI tests on the master branch
Clone the package to your computer, and build with hand or Rstudio. If you need help see here http://biometry.github.io/APES/R/R70-PackageDevelopment.html
In Rstudio, the vignette may not be built per default. You will turn this on in your project options, or run
devtools::build_vignettes("rLPJGUESS")