Home

TOL
Software
Download
Installation
Packages
OTAN

TOL Project
About TOL
Manuals
Development
R-Projects

Other Links
Trac Guide
Wiki Tree
​MMS

Installation

To be revised. Link to TOL+R instructions.

TOL download

Firstly, you should download the installer and run it. See Download.

Additional installation instructions

Some TOL packages (for example, BysInfDiag) uses R calls to obtain some statistic information. For avoid errors when you use these packages we recommend:

Install R

Install the statistical software R. See ​https://www.r-project.org.

Add bin folder to the PATH (for Windows users)

For allow TOL to find R, you should ensure that the R 'bin' folder is in the environment variable PATH.

• R 'bin folder:
  %PROGRAMFILES%\R\R-3.2.2\bin
  (version number may be different)

You can edit the environment variable PATH as usual.

Install some R-packages

Some of the mentioned TOL packages requires from some R-packages: quadprog, coda, Rglpk, slam and mcgibbsit.

To install it, you can run the following sentences in R:

install.packages("quadprog")
install.packages("coda") # imports: lattice
install.packages("mcgibbsit") # depends: coda
install.packages("slam")
install.packages("Rglpk") # depends: slam

External downloads required

In order to use some TOL functions and subsystems it's also needed to install another open source packages or libraries, that are listed in TolAcknowledgements. Linux systems have should have their own package manager (RPM, DEB, etc) for almost all these components. In other case user must install them following owner instructions. For Windows 32 bits systems all them are prebuilt in a transparent way for ussers and are distributed with TOL Binary Setup.

!!!!

All them but R, are linked by TOL and calling is made internally by C or C++ interface. TOL calls R (>=2.6.0) by command line, so it must be installed in your system directly from ​the official R site. Note that R console executable must be in user path environment variable. Afterwords installing R user must install these R packages:

1. ​quadprog : Functions to solve Quadratic Programming Problems.
2. ​coda : Output analysis and diagnostics for MCMC.
3. ​Rglpk : solving large-scale linear programming (LP), mixed integer linear programming (MILP) and other related problems.
4. ​slam : Sparse Lightweight Arrays and Matrices.
5. ​http://cran.r-project.org/web/packages/mcgibbsit/index.html : provides an implementation of Warnes & Raftery's MCGibbsit run-length diagnostic for a set of (not-necessarily independent) MCMC samplers

All them are available as CRAN standard package and can be installed from R-Gui application using menu Package / Install package. You also can run this commands from an R console:

  install.packages("quadprog")
  install.packages("coda")
  install.packages("Rglpk")
  install.packages("slam")
  install.packages("mcgibbsit")

In Debian based LINUX the simplest way is following these steps

  sudo apt-get install r-base
  sudo apt-get install r-cran-quadprog
  sudo apt-get install r-cran-coda
  sudo apt-get install r-cran-rglpk
  sudo apt-get install r-cran-slam
  sudo apt-get install r-cran-mcgibbsit