Mercurial > hg > camir-aes2014

view toolboxes/FullBNT-1.0.7/docs/bnsoftOld.html @ 0:e9a9cd732c1e tip

Find changesets by keywords (author, files, the commit message), revision number or hash, or revset expression.
first hg version after svn
author wolffd
date Tue, 10 Feb 2015 15:05:51 +0000
parents
children
line wrap: on
line source
<head>
<TITLE>Software Packages for Graphical Models / Bayesian Networks</TITLE>
</head>
 
<body>

<h1>
Software Packages for Graphical Models / Bayesian Networks
</h1>
<p>
Written by Kevin Murphy.
<br>
Last updated 31 October  2005.

<h3>Remarks</h3>
<ul>

<li>
A much more detailed comparison of some of these software packages is
available from Appendix B of 
<a href="http://www.csse.monash.edu.au/bai/">Bayesian AI</a>, by 
Ann Nicholson and Kevin Korb.
This appendix is
available
<a
href="http://www.csse.monash.edu.au/bai/book/appendix_b.pdf">here</a>,
and is based on the online comparison below.

<li>
An online <a
href="http://servasi.insa-rouen.fr/enseignement/siteUV/rna/BNT/software.html">French
version</a> of this page is also available (not necessarily up-to-date).

<!--
<li>
Matt Turk has made a
<a href="http://ilab.cs.ucsb.edu/BayesNets/Bayesia_Software.pdf">pdf
version of this page</a> (not necessarily up-to-date) with some extra
comments on the packages.
-->
<!--
<br>
(He also has a
<a href="http://ilab.cs.ucsb.edu/BayesNets">cached copy</a> of several of
the packages listed below, which you can look at if links are down.)
-->

</ul>

<h3>What do the headers in the table mean?</h3>

<ul>
<li> Src = source code included? (N=no) If so, what language?

<li> API = application program interface included?
(N means the program cannot be integrated into your code, i.e., it
must be run as a standalone executable.)

<li> Exec = Executable runs on W = Windows (95/98/NT), U = Unix, M =
Mac, or - = any machine with a compiler.

<li> Cts = are continuous (latent) nodes supported?
G = (conditionally) Gaussians nodes supported analytically,
Cs = continuous nodes supported by sampling,
Cd = continuous nodes supported by discretization,
Cx = continuous nodes supported by some unspecified method,
D = only discrete nodes supported.


<li> GUI = Graphical User Interface included?

<li> Learns parameters?

<li> Learns structure? CI = means uses conditional independency tests

<!--
<li> Sample = sampling methods (e.g., likelihood weighting, MCMC) supported?
-->

<li> Utility = utility and decision nodes (i.e., influence diagrams)
supported?

<li> Free?
0 = free (although possibly only for academic use).
$ = commercial software (although most have free versions
which are restricted in
various ways, e.g., the model size is limited, or models cannot be
saved, or there is no API.)


<li> Undir?
What kind of graphs are supported?
U = only undirected graphs,
D = only directed graphs,
UD = both undirected and directed,
CG = chain graphs (mixed directed/undirected).


<li> Inference = which inference algorithm is used?
jtree = junction tree,
varelim = variable (bucket) elimination,
MH = Metropols Hastings,
G = Gibbs sampling,
IS = importance sampling,
sampling = some other Monte Carlo method,
polytree = Pearl's algorithm restricted to a graph with no cycles, 
none = no inference supported (hence the program is only designed for
structure learning from completely observed data) 

<li> Comments.
If in "quotes", I am quoting the authors at their request.

</ul>



<!--
The following table is automatically generated as follows
bnsoft_to_html.pl < bnsoft.txt > ! soft.html  
Then soft.html is inserted and the last line is tidied up.
-->

<table border>
<tr>
<td> Name
<td> Authors
<td> Src
<td> API
<td> Exec
<td> Cts
<td> GUI
<td> Params
<td> Struct
<td> Utility
<td> Free
<td> Undir
<td> Inference
<td> Comments


<tr>
<td> <!--Name--> <a href=http://www.agenarisk.com > AgenaRisk</a>
<td> <!--Authors-->Agena
<td> <!--Src-->N
<td> <!--API-->Y
<td> <!--Exec-->W,U
<td> <!--Cts-->Cx
<td> <!--GUI-->Y
<td> <!--Params-->Y
<td> <!--Struct-->N
<td> <!--Utility-->N
<td> <!--Free-->$
<td> <!--Undir-->D
<td> <!--Inference-->JTree
<td> <!--Comments-->Simulation by Dynamic discretisation

<tr>
<td> <!--Name--> <a href=http://www.lumina.com > Analytica</a>
<td> <!--Authors-->Lumina
<td> <!--Src-->N
<td> <!--API-->Y
<td> <!--Exec-->W,M
<td> <!--Cts-->G
<td> <!--GUI-->Y
<td> <!--Params-->N
<td> <!--Struct-->N
<td> <!--Utility-->Y
<td> <!--Free-->$
<td> <!--Undir-->D
<td> <!--Inference-->sampling
<td> <!--Comments-->spread sheet compatible

<tr>
<td> <!--Name--> <a href=http://www.cs.duke.edu/~amink/software/banjo/> Banjo</a>
<td> <!--Authors-->Hartemink
<td> <!--Src-->Java
<td> <!--API-->Y
<td> <!--Exec-->W,U,M
<td> <!--Cts-->Cd
<td> <!--GUI-->N
<td> <!--Params-->N
<td> <!--Struct-->Y
<td> <!--Utility-->N
<td> <!--Free-->0
<td> <!--Undir-->D
<td> <!--Inference-->none
<td> <!--Comments--> structure learning of
static or dynamic networks of discrete variables


<tr>
<td><a href=http://www.cs.Helsinki.FI/research/fdk/bassist/> Bassist</a>
<td>        U. Helsinki
<td>            C++
<td>            Y
<td>           U
<td>            G
<td>            N
<td>     Y
<td>     N
<td>        N
<td>            0
<td>      D
<td>      MH
<td>       Generates C++ for MCMC.

<tr>
<td><a href=http://www.cs.Helsinki.FI/research/cosco/Projects/NONE/SW/ > Bayda</a>
<td>        U. Helsinki
<td>            Java
<td>            Y
<td>           WUM
<td>            G
<td>            Y
<td>     Y
<td>     N
<td>        N
<td>            0
<td>      D
<td>      ?
<td>       Bayesian Naive Bayes classifier.

<tr>
<td><a href=http://www.mbfys.kun.nl/snn/Research/bayesbuilder/  >  BayesBuilder</a>
<td>        Nijman (U. Nijmegen)
<td>            N
<td>            N
<td>           W
<td>            D
<td>            Y
<td>     N
<td>     N
<td>        N
<td>         0
<td>      D
<td>      ?
<td>       -

<tr>
<td> <!--Name--> <a
href="http://www.bayesia.com">BayesiaLab</a>
<td> <!--Authors-->Bayesia Ltd
<td> <!--Src-->N
<td> <!--API-->N
<td> <!--Exec-->-
<td> <!--Cts-->Cd
<td> <!--GUI-->Y
<td> <!--Params-->Y
<td> <!--Struct-->Y
<td> <!--Utility-->N
<td> <!--Free-->$
<td> <!--Undir-->CG
<td> <!--Inference-->jtree,G
<td> <!--Comments-->
<!--
Structural learning (association discovery,
classification), Bayesian clustering with analysis report generation,
missing values 
processing, essential graphs, analysis toolbox, adaptive
questionnaires, dynamic models
-->
Structural learning, adaptive
questionnaires, dynamic models




<tr>

<td><a href=http://www.bayesware.com>Bayesware Discoverer</a>
<td>        Bayesware
<td>            N
<td>            N
<td>           WUM
<td>            Cd
<td>            Y
<td>     Y
<td>     Y
<td>        N
<td>            $
<td>      D
<td>      ?
<td>       Uses bound and collapse for learning with missing data.

<tr>
<td><a href=http://B-Course.hiit.fi/> B-course</a>
<td>        U.  Helsinki
<td>            N
<td>            N
<td>           WUM
<td>            Cd
<td>            Y
<td>     Y
<td>     Y
<td>        N
<td>            0
<td>      D
<td>      ?
<td>       Runs on their server: view results using a web browser.

<!--
<tr>
<td><a href="http://www.aist.go.jp/ETL/etl/suri/motomura/BN/bn-java.html">Bayonnet</a>
<td>        Motomura (ETL)
<td>            Java
<td>            Y
<td>           WUM
<td>            NN
<td>            Cx
<td>     Y
<td>     N
<td>        N
<td>            0
<td>      D
<td>      ?
<td>       For learning, represents BN as a neural net.
-->

<tr>
<td><a href=http://www.cs.ualberta.ca/~jcheng/bnpc.htm  >  Belief net power constructor</a>
<td>        Cheng (U.Alberta)
<td>            N
<td>            W
<td>           W
<td>            D
<td>            Y
<td>     Y
<td>     CI
<td>        N
<td>            0
<td>      D
<td>      ?
<td>       -

<tr>
<td><a href=http://www.cs.berkeley.edu/~murphyk/Bayes/bnt.html  >  BNT</a>
<td>        Murphy (U.C.Berkeley)
<td>            Matlab/C
<td>            Y
<td>           WUM
<td>            G
<td>            N
<td>     Y
<td>     Y
<td>        Y
<td>            0
<td>      D,U
<td>      Many
<td>       Also handles dynamic models, like HMMs and Kalman filters.


<tr>
<td> <!--Name--> <a href="http://bndev.sourceforge.net/">BNJ</a>
<td> <!--Authors-->Hsu (Kansas)
<td> <!--Src-->Java
<td> <!--API-->-
<td> <!--Exec-->-
<td> <!--Cts-->D
<td> <!--GUI-->Y
<td> <!--Params-->N
<td> <!--Struct-->Y
<td> <!--Utility-->N
<td> <!--Free-->0
<td> <!--Undir-->D
<td> <!--Inference-->jtree, IS
<td> <!--Comments-->-

<!--
<tr>
<td><a href=http://94.parkview-court.net/project/  >  BN Toolkit</a>
<td>        Gowans (Imperial)
<td>            Visual Basic
<td>            Y
<td>           W
<td>            D
<td>            Y
<td>     N
<td>     Y
<td>        N
<td>            0
<td>      D
<td>      Polytree
<td>       Parser and GUI for the XML-BIF format.
-->

<tr>
<td><a href=http://www.ics.uci.edu/~irinar  >  BucketElim</a>
<td>        Rish (U.C.Irvine)
<td>            C++
<td>            Y
<td>           WU
<td>            D
<td>            N
<td>     N
<td>     N
<td>        N
<td>            0
<td>      D
<td>      Varelim
<td>       -

<tr>
<td><a href=http://www.mrc-bsu.cam.ac.uk/bugs  >  BUGS</a>
<td>        MRC/Imperial College
<td>            N
<td>            N
<td>           WU
<td>            Cs
<td>            W
<td>     Y
<td>     N
<td>        N
<td>            0
<td>      D
<td>      Gibbs
<td>       -

<tr>
<td><a href=http://www.data-digest.com  >  Business Navigator 5</a>
<td>        Data Digest Corp
<td>            N
<td>            N
<td>           W
<td>            Cd
<td>            Y
<td>     Y
<td>     Y
<td>        N
<td>            $
<td>      D
<td>      Jtree
<td>       -

<tr>
<td><a href=http://www-pcd.stanford.edu/cousins/caben-1.1.tar.gz  >  CABeN</a>
<td>        Cousins et al. (Wash. U.)
<td>            C
<td>            Y
<td>           WU
<td>            D
<td>            N
<td>     N
<td>     N
<td>        N
<td>            0
<td>      D
<td>      5 Sampling methods
<td>       -

<tr>
<td> <!--Name--> <a href=
http://discover1.mc.vanderbilt.edu/discover/public/>Causal discoverer</a>
<td> <!--Authors-->Vanderbilt
<td> <!--Src-->N
<td> <!--API-->N
<td> <!--Exec-->W
<td> <!--Cts-->-
<td> <!--GUI-->-
<td> <!--Params-->N
<td> <!--Struct-->Y
<td> <!--Utility-->N
<td> <!--Free-->0
<td> <!--Undir-->D
<td> <!--Inference-->-
<td> <!--Comments-->structure learning only


<tr>
<td><a href=http://www.math.auc.dk/~jhb/CoCo/information.html  >  CoCo+Xlisp</a>
<td>        Badsberg (U. Aalborg)
<td>            C/lisp
<td>            Y
<td>           U
<td>            D
<td>            Y
<td>     Y
<td>     CI
<td>        N
<td>            0
<td>      U
<td>      Jtree
<td>       Designed for contingency tables.

<tr>
<td><a href=http://www.cs.ubc.ca/labs/lci/CIspace/  >  CIspace</a>
<td>        Poole et al. (UBC)
<td>            Java
<td>            N
<td>           WU
<td>            D
<td>            Y
<td>     N
<td>     N
<td>        N
<td>            0
<td>      D
<td>      Varelim
<td>       -

<tr>
<td> <!--Name--> <a href="http://www.robots.ox.ac.uk/~parg/software.html">DBNbox</a>
<td> <!--Authors-->Roberts et al
<td> <!--Src-->Matlab
<td> <!--API-->-
<td> <!--Exec-->-
<td> <!--Cts-->Y
<td> <!--GUI-->N
<td> <!--Params-->Y
<td> <!--Struct-->N
<td> <!--Utility-->N

<td> <!--Free-->Y
<td> <!--Undir-->D
<td> <!--Inference-->Various
<td> <!--Comments-->DBNs

<tr>
<td> <!--Name--><a href=http://www.math.auc.dk/novo/deal>Deal</a>
<td> <!--Authors-->Bottcher et al
<td> <!--Src-->R
<td> <!--API-->-
<td> <!--Exec-->-
<td> <!--Cts-->G
<td> <!--GUI-->Y
<td> <!--Params-->Y
<td> <!--Struct-->Y
<td> <!--Utility-->N
<td> <!--Free-->0
<td> <!--Undir-->D
<td> <!--Inference-->None
<td> <!--Comments-->
Structure learning.

<tr>
<td><a href=http://www.deriveit.com >DeriveIt</a>
<td> <!--Authors-->DeriveIt LLC
<td> <!--Src-->N
<td> <!--API-->-
<td> <!--Exec-->-
<td> <!--Cts-->?
<td> <!--GUI-->?
<td> <!--Params-->Y
<td> <!--Struct-->Y
<td> <!--Utility-->?
<td> <!--Free-->$
<td> <!--Undir-->D
<td> <!--Inference-->Jtree
<td> <!--Comments-->
Exploits local structure in CPDs.


<tr>
<td><a href=http://www.noeticsystems.com>Ergo</>
<td> <!--Authors-->Noetic systems
<td> <!--Src-->N
<td> <!--API-->Y
<td> <!--Exec-->W,M
<td> <!--Cts-->D
<td> <!--GUI-->Y
<td> <!--Params-->N
<td> <!--Struct-->N
<td> <!--Utility-->N
<td> <!--Free-->$
<td> <!--Undir-->D
<td> <!--Inference-->jtree
<td> <!--Comments-->-

<!--
<tr>
<td><a href=http://www.ens-lyon.fr/~jnarboux/Floue/index.html>FLoUE/BIFtoN</a>
<td>        ENS Lyon
<td>            Java
<td>            Y
<td>           WUM
<td>           D
<td>            N
<td>     N
<td>     N
<td>        N
<td>            0
<td>      D
<td>      Jtree
<td>       -
-->

<tr>
<td><a href=http://www.staff.ncl.ac.uk/d.j.wilkinson/software/gdagsim/  >  GDAGsim</a>
<td>        Wilkinson (U. Newcastle)
<td>            C
<td>            Y
<td>           WUM
<td>            G
<td>            N
<td>     N
<td>     N
<td>        N
<td>            0
<td>      D
<td>      Exact
<td>    Bayesian analysis of large linear Gaussian directed models.


<tr>
<td><a href=http://www2.sis.pitt.edu/~genie  >  Genie</a>
<td>        U. Pittsburgh
<td>            N
<td>            WU
<td>           WU
<td>            D
<td>            W
<td>     N
<td>     N
<td>        Y
<td>            0
<td>      D
<td>      Jtree
<td>       -

<tr>
<td><a href=http://www.math.ntnu.no/~hrue/GMRFsim/  >  GMRFsim</a>
<td>        Rue (U. Trondheim)
<td>            C
<td>            Y
<td>           WUM
<td>            G
<td>            N
<td>     N
<td>     N
<td>        N
<td>            0
<td>      U
<td>      MCMC
<td>       Bayesian analysis of large linear Gaussian undirected models.


<tr>
<td> <!--Name--> <a href=http://ssli.ee.washington.edu/~bilmes/gmtk/>GMTk</a>
<td> <!--Authors-->Bilmes (UW), Zweig (IBM)
<td> <!--Src-->N
<td> <!--API-->Y
<td> <!--Exec-->U
<td> <!--Cts-->D
<td> <!--GUI-->N
<td> <!--Params-->Y
<td> <!--Struct-->Y
<td> <!--Utility-->N
<td> <!--Free-->0
<td> <!--Undir-->D
<td> <!--Inference-->Jtree
<td> <!--Comments-->
Designed for speech recognition.

<tr>
<td> <!--Name--> <a href=http://www.r-project.org/gR>gR</a>
<td> <!--Authors-->Lauritzen et al.
<td> <!--Src-->R
<td> <!--API-->-
<td> <!--Exec-->-
<td> <!--Cts-->-
<td> <!--GUI-->-
<td> <!--Params-->-
<td> <!--Struct-->-
<td> <!--Utility-->-
<td> <!--Free-->0
<td> <!--Undir-->-
<td> <!--Inference-->-
<td> <!--Comments-->Currently vaporware


<tr>
<td> <!--Name--> <a href=http://www.stats.bris.ac.uk/~peter/Grappa/>Grappa</a>
<td> <!--Authors-->Green (Bristol)
<td> <!--Src-->R
<td> <!--API-->-
<td> <!--Exec-->-
<td> <!--Cts-->D
<td> <!--GUI-->N
<td> <!--Params-->N)
<td> <!--Struct-->N
<td> <!--Utility-->N
<td> <!--Free-->0
<td> <!--Undir-->D
<td> <!--Inference-->Jtree
<td> <!--Comments-->-


<tr>
<td><a href=http://www.hugin.com  >  Hugin Expert</a>
<td>        Hugin
<td>            N
<td>            Y
<td>           W
<td>            G
<td>            W
<td>     Y
<td>     CI
<td>        Y
<td>            $
<td>      CG
<td>      Jtree
<td>       -


<tr>
<td> <!--Name--> <a href=http://www.warnes.net/GregsSoftwareLinks/index_html?Tab=MCMC>Hydra</a>
<td> <!--Authors-->Warnes (U.Wash.)
<td> <!--Src-->Java
<td> <!--API-->-
<td> <!--Exec-->-
<td> <!--Cts-->Cs
<td> <!--GUI-->Y
<td> <!--Params-->Y
<td> <!--Struct-->N
<td> <!--Utility-->N
<td> <!--Free-->0
<td> <!--Undir-->U,D
<td> <!--Inference-->MCMC
<td> <!--Comments-->-


<tr>
<!-- http://www.rpal.rockwell.com/ideal.html  -->
<td><a href=http://yoda.cis.temple.edu:8080/ideal>Ideal</a>
<td>        Rockwell
<td>            Lisp
<td>            Y
<td>           WUM
<td>            D
<td>            Y
<td>     N
<td>     N
<td>        Y
<td>            0
<td>      D
<td>      Jtree
<td>       GUI requires Allegro Lisp.

<tr>
<td><a href=http://www.cs.cmu.edu/~javabayes/Home/  >  Java Bayes</a>
<td>        Cozman (CMU)
<td>            Java
<td>            Y
<td>           WUM
<td>            D
<td>            Y
<td>     N
<td>     N
<td>        Y
<td>            0
<td>      D
<td>      Varelim, jtree
<td>       -


<tr>
<td> <!--Name--><a href="http://www.codeas.com/kbaseai.php" >KBaseAI</a>
<td> <!--Authors-->Codeas
<td> <!--Src-->N
<td> <!--API-->Y
<td> <!--Exec-->W,U
<td> <!--Cts-->D
<td> <!--GUI-->N
<td> <!--Params-->N
<td> <!--Struct-->N
<td> <!--Utility-->N
<td> <!--Free-->$
<td> <!--Undir-->D
<td> <!--Inference-->varelim
<td> <!--Comments-->client/server architecture, multiple users, access
control, query language


<tr>
<td> <!--Name--> <a href=http://www.cs.huji.ac.il/labs/compbio/LibB>LibB</a>
<td> <!--Authors-->Friedman (Hebrew U)
<td> <!--Src-->N
<td> <!--API-->Y
<td> <!--Exec-->W
<td> <!--Cts-->D
<td> <!--GUI-->N
<td> <!--Params-->Y
<td> <!--Struct-->Y
<td> <!--Utility-->N
<td> <!--Free-->0
<td> <!--Undir-->D
<td> <!--Inference-->none
<td> <!--Comments-->
Structure learning

<tr>
<td><a href=http://www.hypergraph.dk/  >  MIM</a>
<td>        HyperGraph Software
<td>            N
<td>            N
<td>           W
<td>            G
<td>            Y
<td>     Y
<td>     Y
<td>        N
<td>            $
<td>      CG
<td>      Jtree
<td>       Up to 52 variables.

<tr>
<td><a href=http://research.microsoft.com/adapt/MSBNx/  >  MSBNx</a>
<td>        Microsoft
<td>            N
<td>            Y
<td>           W
<td>            D
<td>            W
<td>     N
<td>     N
<td>        Y
<td>            0
<td>      D
<td>      Jtree
<td>       -

<tr>
<td><a href=http://www.norsys.com  >  Netica</a>
<td>        Norsys
<td>            N
<td>            WUM
<td>           W
<td>            G
<td>            W
<td>     Y
<td>     N
<td>        Y
<td>            $
<td>      D
<td>      jtree
<td>       -

<tr>
<td> <!--Name--> <a href="http://www.autonlab.org/autonweb/showSoftware/149/">Optimal
Reinsertion</a>
<td> <!--Authors-->Moore, Wong (CMU)
<td> <!--Src-->N
<td> <!--API-->N
<td> <!--Exec-->W,U
<td> <!--Cts-->D
<td> <!--GUI-->N
<td> <!--Params-->Y
<td> <!--Struct-->Y
<td> <!--Utility-->N
<td> <!--Free-->0
<td> <!--Undir-->D
<td> <!--Inference-->none
<td> <!--Comments-->structure learning


<tr>
<td> <!--Name--> <a href="http://people.bu.edu/vladimir/pmt/index.html">PMT</a>
<td> <!--Authors-->Pavlovic (BU)
<td> <!--Src-->Matlab/C
<td> <!--API-->-
<td> <!--Exec-->-
<td> <!--Cts-->D
<td> <!--GUI-->N
<td> <!--Params-->Y
<td> <!--Struct-->N
<td> <!--Utility-->N
<td> <!--Free-->0
<td> <!--Undir-->D
<td> <!--Inference-->special purpose
<td> <!--Comments-->-



<tr>
<td> <!--Name--> <a href=http://www.intel.com/research/mrl/pnl/>PNL</a>
<td> <!--Authors-->Eruhimov (Intel)
<td> <!--Src-->C++
<td> <!--API-->-
<td> <!--Exec-->-
<td> <!--Cts-->D
<td> <!--GUI-->N
<td> <!--Params-->Y
<td> <!--Struct-->Y
<td> <!--Utility-->N
<td> <!--Free-->0
<td> <!--Undir-->U,D
<td> <!--Inference-->Jtree
<td> <!--Comments-->
A C++ version of BNT; will be released 12/03.


<tr>
<td><a href=http://iridia.ulb.ac.be/pulcinella/Welcome.html  >  Pulcinella</a>
<td>        IRIDIA
<td>            Lisp
<td>            Y
<td>           WUM
<td>            D
<td>            Y
<td>     N
<td>     N
<td>        N
<td>            0
<td>      D
<td>      ?
<td>       Uses valuation systems for non-probabilistic calculi.

<tr>
<!-- http://civil.colorado.edu/~dodier/sonero-mirror/index.html  -->
<td><a href=http://sourceforge.net/projects/riso>RISO</a>
<td>        Dodier (U.Colorado)
<td>            Java
<td>            Y
<td>           WUM
<td>            G
<td>            Y
<td>     N
<td>     N
<td>        N
<td>            0
<td>      D
<td>      Polytree
<td>       Distributed implementation.



<tr>
<td> <!--Name--> <a href=http://reasoning.cs.ucla.edu/samiam/> Sam Iam</a>
<td> <!--Authors-->Darwiche (UCLA)
<td> <!--Src-->N
<td> <!--API-->N ?
<td> <!--Exec-->WU ? (Java executable)
<td> <!--Cts-->G ?
<td> <!--GUI-->Y
<td> <!--Params-->Y
<td> <!--Struct-->N ?
<td> <!--Utility-->Y
<td> <!--Free-->0
<td> <!--Undir-->D
<td> <!--Inference-->Recursive conditioning
<td> <!--Comments-->Also does sensitivity Analysis


<tr>
<td><a href=http://www.phil.cmu.edu/tetrad/>Tetrad</a>
<td>        CMU
<td>            N
<td>            N
<td>           WU
<td>            G
<td>            N
<td>     Y
<td>     CI
<td>        N
<td>            0
<td>      U,D
<td>      None
<td>       -


<tr>
<td> <!--Name--> <a href="https://sourceforge.net/projects/unbbayes/">UnBBayes</a>
<td> <!--Authors-->?
<td> <!--Src-->Java
<td> <!--API-->-
<td> <!--Exec-->-
<td> <!--Cts-->D
<td> <!--GUI-->Y
<td> <!--Params-->N
<td> <!--Struct-->Y
<td> <!--Utility-->N
<td> <!--Free-->0
<td> <!--Undir-->D
<td> <!--Inference-->jtree
<td> <!--Comments-->K2 for struct learning


<tr>
<td><a href=http://www.inference.phy.cam.ac.uk/jmw39/>Vibes</a>
<td>        Winn & Bishop (U. Cambridge)
<td>            Java
<td>            Y
<td>           WU
<td>            Cx
<td>            Y
<td>     Y
<td>     N
<td>        N
<td>            0
<td>      D
<td>      Variational
<td>    
Not yet available.


<tr>
<td><a href=http://snowhite.cis.uoguelph.ca/faculty_info/yxiang/ww3/>  Web Weaver</a>
<td>        Xiang (U.Regina)
<td>            Java
<td>            Y
<td>           WUM
<td>            D
<td>            Y
<td>     N
<td>     N
<td>        Y
<td>            0
<td>      D
<td>      ?
<td>       -

<tr>
<td><a href=http://research.microsoft.com/~dmax/WinMine/tooldoc.htm  >  WinMine</a>
<td>        Microsoft
<td>            N
<td>            N
<td>           W
<td>            Cx
<td>            Y
<td>     Y
<td>     Y
<td>        N
<td>            0
<td>      U,D
<td>      None
<td>       Learns BN or dependency net structure.

<tr>
<td><a href=http://www.staff.city.ac.uk/~rgc/webpages/xbpage.html >  XBAIES 2.0</a>
<td>        Cowell (City U.)
<td>            N
<td>            N
<td>           W
<td>            G
<td>            Y
<td>     Y
<td>     N
<td>        Y
<td>            0
<td>      CG
<td>      Jtree
<td>       -

<tr>

</table>


<p>
<h2>Other sites related to (software for) graphical models</h2>
<ul>


<li>
<a
href="http://directory.google.com/Top/Computers/Artificial_Intelligence/Belief_Networks/Software/">
Google's list</a> of Bayes net software.

<!--
<li> <a href="http://www.sis.pitt.edu/~dsl/da-software.html">
Decision analysis programs</a>, list maintained by 
Marek Druzdzel.
-->

<li> <a href="http://www.cs.cmu.edu/~lorens/papers/mscthesis.html">
Comparison of decision analysis software packages</a> by
Håkan L. Younes. The emphasis is on real-time decision making.

<li> <a href="http://bayes.stat.washington.edu/almond/belief.html">
Older belief net programs</a> (c 1996), a list created (but no longer maintained)
by Russ Almond. 

<li>
<a
href="http://www.cs.huji.ac.il/labs/compbio/Repository">
Repository of Bayes nets</a>.

</document>