fetal liver image, protein structure, galaxy image, Bayes net for forensic DNA analysis, gene expression data Concorde final flight
Mathematics homeStatistics homeResearchPeopleSeminarsPostgrad opportunities

Statistics SuSTaIn Peter Green Research Collaborators Papers Software Books BGX Funded research projects Teaching > Links Personal APTS Complexity science Royal Statistical Society International Society for Bayesian Analysis Royal Society


PJG Home / Software

Grappa

-- a suite of functions in R for probability propagation in discrete graphical models.
If you download the R functions or the user guide, please enter your email address and any comments in this form, and press "Submit". I can then keep you informed of any updates.

Your email address:
Please type your message here:


  • R source code, version 0.3 [13/12/09] can be used safely with 0.2 Fortran code, DLL and examples
  • R source code, version 0.2 [21/11/05] superseded
  • Fortran source code for key routines, version 0.2 [21/11/05]
  • Windows DLL from this Fortran source code, version 0.2 [21/11/05]
  • User guide in PDF format [21/11/05]
  • Changes between versions [13/12/09]
  • R package for Windows and Linux for version 0.2 [21/11/05] (provided by Iago Mosqueira; please direct any queries to imosqueira (at) suk.azti.es)
  • Input file of examples [12/01/03] and output [9/12/03] (run under R version 1.8.1).
  • A PowerPoint show about probability propagation
  • Handout for PowerPoint show (PDF format)
  • Home page for Hugin
  • R can be downloaded from here

Download instructions: as a minimum, download the R source into your current working directory, and type source('Grappa.R') within R. To test, run the examples, using source or by copy-paste. To improve performance, download and install the Fortran code, or if you are using Windows, the DLL version. See Section 3.4 of the User guide.
Further updates can be expected at random.

Grappa is free of charge for educational and non-commercial research purposes.
Please report any problems with the code to the author, Peter Green, via Email link