RHugin Install Instructions for Mac OS X

Last updated by Kjell Konis on 2017-02-02.


Preliminaries

The RHugin package provides an R API for the HUGIN Decision Engine (HDE): both the HDE and R must be installed on the computer in order for the RHugin package to function. Since RHugin is distributed as a source package, the command line developer tools are also required.

Install HUGIN and the HUGIN Decision Engine (HDE)

The HUGIN Decision Engine is commercial software available from HUGIN EXPERT.

http://www.hugin.com

The products HUGIN Developer (commercial use) and HUGIN Researcher (academic use) include the HUGIN Decision Engine. The HDE is also included with the free demo HUGIN Lite (the only permitted use of the free demo is to evaluate HUGIN). HUGIN is distributed on a disk image (i.e., a .dmg file). Mount the disk image by double clicking on it, navigate to the mounted disk, and install the HDE by dragging the HDE folder and dropping it onto the Applications alias. You may also want to install the other components (HUGIN GUI, documentation, etc.) but these are not required for RHugin.

Install R

Just in case you don’t already have R installed… R is available from the CRAN repository.

https://cloud.r-project.org/

The installer for the most recent version of R can be found by following the Download R for (Mac) OS X link.

Install Command Line Developer Tools

Start R and run the following command to install the command line developer tools.

system2("xcode-select", "--install")

A dialog box will pop up and prompt you to install the command line developer tools. Click Install (note: this command must be run by a user with administrative privileges).

If you get the following error

xcode-select: error: command line tools are already installed, use "Software Update" to install updates

the command line developer tools are already installed and you could have skipped this section.


RHugin Package Depedencies

The RHugin package has dependencies on the graph and Rgraphviz packages. These packages are part of Bioconductor and can be installed by running the following two R commands.

source("https://bioconductor.org/biocLite.R")
biocLite(c("graph", "Rgraphviz"))

Install the RHugin Package

First, set the HUGINHOME environment variable so that RHugin can find the HUGIN Decision Engine (HDE). The value (i.e., the "/Applications/HDE8.4-x64" in the code below) should be the full path to the HDE folder that you installed in the Preliminaries section.

Sys.setenv(HUGINHOME = "/Applications/HDE8.4-x64")

Then, install the RHugin package from RHugin’s R-Forge repository with the following R command.

install.packages("RHugin", repos = "http://rhugin.r-forge.r-project.org",
                 type = "source", INSTALL_opts="--no-multiarch")

Quick Test

To check that everything went as planned run the following code to plot one of the example networks provided by HUGIN.

library(RHugin)
demo("asia", echo = FALSE)
plot(asia)


Happy HUGINing with RHugin!




Copyright 2017 Kjell Konis