admin:rui [Promethee]

[R] front-end installation guide (1.1 version)

As a new feature of Promethee 1.1, it is now possible to call Promethee as a [R] function. This usage is highly valuable for coupling codes (like a composition of functions), testing new algorithms, or generally, using an external code (available in Promethee environment) like another R function.
The requirement to use Promethee within [R], is firstly to have both Promethee and [R] installed on the same machine, and then to properly install the rJava package in [R] (see http://www.rforge.net/rJava/).

Checking the installation

Once [R] is started,

load the promethee.R script available in the Promethee installation directory,
initialize Promethee with Promethee.init(…) function,

then use shell functions (Promethee.run(…) and Promethee.design(…)) like another R function:

R version 2.14.1 (2011-12-22)
Copyright (C) 2011 The R Foundation for Statistical Computing
ISBN 3-900051-07-0
Platform: x86_64-pc-linux-gnu (64-bit)
...
 
 > PROMETHEE_HOME="/opt/Promethee"
 > source(file.path(PROMETHEE_HOME,"promethee.R"))
 > Promethee.init(PROMETHEE_HOME)
Loading required package: rJava
Java Java(TM) SE Runtime Environment
 version 1.6.0_26
 from path /usr/lib/jvm/java-6-sun-1.6.0.26/jre
Promethee 1.1-0 <build 09/01/2012 17:07>
Reading quota configuration
Initializing math. engine
Supported models: [... MCNP_5, ...]
Maximum load: 10 computers
Supported designs: [...]
 > Promethee.run(model = "MCNP_5",input.files = file.path(PROMETHEE_HOME,"samples","godiva.p"),input.design = list(r=c(7.0,8.0,9.0)))
$mean_keff
[1] 0.82429 0.92694 1.02363
 
$sigma_keff
[1] 0.00054 0.00057 0.00061

More information and examples are available as comments in the promethee.R file.
It is also recommended to use the RStudio front-end environment (see http://www.rstudio.org) in order to make the most of R in desktop usage, especially when using Promethee interface.

FAQ

What is the value added of the [R] front-end compared to the regular GUI ?

Why do you recommend to use Rstudio environment ?

Can I use the web based version of RStudio ?

I understand a “Promethee code” call is just like a R function call. Ok, but what about parallelization of function calls (mclapply, foreach, …) ?

What operating systems are supported by this front-end ?