1 libDAI - A free/open source C++ library for Discrete Approximate Inference
3 -------------------------------------------------------------------------------
6 Date: May 12, 2010, or later
7 See also: http://www.libdai.org
9 -------------------------------------------------------------------------------
13 libDAI is free software; you can redistribute it and/or modify it under the
14 terms of the GNU General Public License as published by the Free Software
15 Foundation; either version 2 of the License, or (at your option) any later
18 libDAI is distributed in the hope that it will be useful, but WITHOUT ANY
19 WARRANTY; without even the implied warranty of MERCHANTABILITY or FITNESS FOR A
20 PARTICULAR PURPOSE. See the GNU General Public License for more details.
22 You should have received a copy of the GNU General Public License
23 along with libDAI in the file COPYING. If not, see http://www.gnu.org/licenses/
25 -------------------------------------------------------------------------------
29 If you write a scientific paper describing research that made substantive use
30 of this program, please cite the software appropriately, by mentioning the
31 fashion in which this software was used, including the version number.
33 An appropriate citation would be:
35 Joris M. Mooij et al. (2010) "libDAI 0.2.5: A free/open source C++ library for
36 Discrete Approximate Inference", http://www.libdai.org
40 @misc{mooij2010libdai,
41 author = "Joris M. Mooij et al.",
42 title = "lib{DAI} 0.2.5: A free/open source {C}++ library for {D}iscrete {A}pproximate {I}nference",
43 howpublished = "http://www.libdai.org/",
48 Moreover, as a personal note, I would appreciate it to be informed about any
49 publications using libDAI at joris dot mooij at libdai dot org.
51 -------------------------------------------------------------------------------
55 libDAI is a free/open source C++ library (licensed under GPL 2+) that provides
56 implementations of various (approximate) inference methods for discrete
57 graphical models. libDAI supports arbitrary factor graphs with discrete
58 variables; this includes discrete Markov Random Fields and Bayesian Networks.
60 The library is targeted at researchers. To be able to use the library, a good
61 understanding of graphical models is needed.
63 The best way to use libDAI is by writing C++ code that invokes the library; in
64 addition, part of the functionality is accessibly by using the
66 * command line interface
67 * (limited) MatLab interface
68 * (experimental) python interface
69 * (experimental) octave interface.
71 libDAI can be used to implement novel (approximate) inference algorithms and to
72 easily compare the accuracy and performance with existing algorithms that have
73 been implemented already.
77 Currently, libDAI supports the following (approximate) inference methods:
79 * Exact inference by brute force enumeration;
80 * Exact inference by junction-tree methods;
82 * Loopy Belief Propagation [KFL01];
83 * Fractional Belief Propagation [WiH03];
84 * Tree-Reweighted Belief Propagation [WJW03];
85 * Tree Expectation Propagation [MiQ04];
86 * Generalized Belief Propagation [YFW05];
87 * Double-loop GBP [HAK03];
88 * Various variants of Loop Corrected Belief Propagation [MoK07, MoR05];
90 * Conditioned Belief Propagation [EaG09];
91 * Decimation algorithm.
93 These inference methods can be used to calculate partition sums, marginals over
94 subsets of variables, and MAP states (the joint state of variables that has
97 In addition, libDAI supports parameter learning of conditional probability
98 tables by Expectation Maximization.
102 libDAI is not intended to be a complete package for approximate inference.
103 Instead, it should be considered as an "inference engine", providing various
104 inference methods. In particular, it contains no GUI, currently only supports
105 its own file format for input and output (although support for standard file
106 formats may be added later), and provides very limited visualization
107 functionalities. The only learning method supported currently is Expectation
108 Maximization (or Maximum Likelihood if no data is missing) for learning factor
113 In my opinion, the lack of open source "reference" implementations hampers
114 progress in research on approximate inference. Methods differ widely in terms
115 of quality and performance characteristics, which also depend in different ways
116 on various properties of the graphical models. Finding the best approximate
117 inference method for a particular application therefore often requires
118 empirical comparisons. However, implementing and debugging these methods takes
119 a lot of time which could otherwise be spent on research. I hope that this code
120 will aid researchers to be able to easily compare various (existing as well as
121 new) approximate inference methods, in this way accelerating research and
122 stimulating real-world applications of approximate inference.
126 Because libDAI is implemented in C++, it is very fast compared with
127 implementations in MatLab (a factor 1000 faster is not uncommon). libDAI does
128 provide a (limited) MatLab interface for easy integration with MatLab. It also
129 provides a command line interface and experimental python and octave interfaces
130 (thanks to Patrick Pletscher).
134 The code has been developed under Debian GNU/Linux with the GCC compiler suite.
135 libDAI compiles successfully with g++ versions 3.4 up to 4.4.
137 libDAI has also been successfully compiled with MS Visual Studio 2008 under
138 Windows (but not all build targets are supported yet) and with Cygwin under
141 Finally, libDAI has been compiled successfully on MacOS X.
145 The libDAI sources and documentation can be downloaded from the libDAI website:
146 http://www.libdai.org.
150 The Google group "libDAI" (http://groups.google.com/group/libdai) can be used
151 for getting support and discussing development issues.
153 -------------------------------------------------------------------------------
155 Building libDAI under UNIX variants (Linux / Cygwin / Mac OS X)
161 * a recent version of gcc (at least version 3.4)
163 * recent boost C++ libraries (at least version 1.37; however, version 1.37
164 shipped with Ubuntu 9.04 is known not to work)
165 * doxygen (only for building the documentation)
166 * graphviz (only for using some of the libDAI command line utilities)
167 * CImg library (only for building the image segmentation example)
169 On Debian/Ubuntu, you can easily install the required packages with a single
172 apt-get install g++ make doxygen graphviz libboost-dev libboost-graph-dev libboost-program-options-dev libboost-test-dev cimg-dev
174 (root permissions needed).
176 On Mac OS X (10.4 is known to work), these packages can be installed easily via
177 MacPorts. If MacPorts is not already installed, install it according to the
178 instructions at http://www.macports.org/. Then, a simple
180 sudo port install gmake boost doxygen graphviz
182 should be enough to install everything that is needed.
184 On Cygwin, the prebuilt Cygwin package boost-1.33.1-x is known not to work. You
185 can however obtain the latest boost version (you need at least 1.37.0) from
186 http://www.boost.org/ and build it as described in the next subsection.
188 Building boost under Cygwin
190 * Download the latest boost libraries from http://www.boost.org
191 * Build the required boost libraries using:
193 ./bootstrap.sh --with-libraries=program_options,math,graph,test --prefix=/boost_root/
196 * In order to use dynamic linking, the boost .dll's should be somewhere in
197 the path. This can be achieved by a command like:
199 export PATH=$PATH:/boost_root/stage/lib
203 To build the libDAI source, first copy a template Makefile.* to Makefile.conf
204 (for example, copy Makefile.LINUX to Makefile.conf if you use GNU/Linux). Then,
205 edit the Makefile.conf template to adapt it to your local setup. Especially
206 directories may differ from system to system. Platform independent build
207 options can be set in Makefile.ALL. Finally, run
211 The build includes a regression test, which may take a while to complete.
213 If the build is successful, you can test the example program:
215 examples/example tests/alarm.fg
217 or the more extensive test program:
219 tests/testdai --aliases tests/aliases.conf --filename tests/alarm.fg --methods JTREE_HUGIN BP_SEQMAX
221 -------------------------------------------------------------------------------
223 Building libDAI under Windows
229 * A recent version of MicroSoft Visual Studio (2008 is known to work)
230 * recent boost C++ libraries (version 1.37 or higher)
231 * GNU make (can be obtained from http://gnuwin32.sourceforge.net)
232 * CImg library (only for building the image segmentation example)
234 For the regression test, you need:
236 * GNU diff, GNU sed (can be obtained from http://gnuwin32.sourceforge.net)
238 Building boost under Windows
240 Because building boost under Windows is tricky, I provide some guidance here.
242 * Download the boost zip file from http://www.boost.org/users/download and
244 * Download the bjam executable from http://www.boost.org/users/download and
245 unpack it somewhere else.
246 * Download Boost.Build (v2) from http://www.boost.org/docs/tools/build/
247 index.html and unpack it yet somewhere else.
248 * Edit the file boost-build.jam in the main boost directory to change the
249 BOOST_BUILD directory to the place where you put Boost.Build (use UNIX /
250 instead of Windows \ in pathnames).
251 * Copy the bjam.exe executable into the main boost directory. Now if you
252 issue "bjam --version" you should get a version and no errors. Issueing
253 "bjam --show-libraries" will show the libraries that will be built.
254 * The following command builds the boost libraries that are relevant for
257 bjam --with-graph --with-math --with-program_options --with-test link=static runtime-link=shared
261 To build the source, copy Makefile.WINDOWS to Makefile.conf. Then, edit
262 Makefile.conf to adapt it to your local setup. Platform independent build
263 options can be set in Makefile.ALL. Finally, run (from the command line)
267 The build includes a regression test, which may take a while to complete.
269 If the build is successful, you can test the example program:
271 examples\example tests\alarm.fg
273 or the more extensive test program:
275 tests\testdai --aliases tests\aliases.conf --filename tests\alarm.fg --methods JTREE_HUGIN BP_SEQMAX
277 -------------------------------------------------------------------------------
279 Building the libDAI MatLab interface
284 * The platform-dependent requirements described above
286 First, you need to build the libDAI source as described above for your
287 platform. By default, the MatLab interface is disabled, so before compiling the
288 source, you have to enable it in Makefile.ALL by setting
292 Also, you have to configure the MatLab-specific parts of Makefile.conf to match
293 your system (in particular, the Makefile variables ME, MATLABDIR and MEX). The
294 MEX file extension depends on your platform; for a 64-bit linux x86_64 system
295 this would be "ME=.mexa64", for a 32-bit linux x86 system "ME=.mexglx". If you
296 are unsure about your MEX file extension: it needs to be the same as what the
297 MatLab command "mexext" returns. The required MEX files are built by issuing
301 from the command line. The MatLab interface is much less powerful than using
302 libDAI from C++. There are two reasons for this: (i) it is boring to write MEX
303 files; (ii) the large performance penalty paid when large data structures (like
304 factor graphs) have to be converted between their native C++ data structure to
305 something that MatLab understands.
307 A simple example of how to use the MatLab interface is the following (entered
308 at the MatLab prompt), which performs exact inference by the junction tree
309 algorithm and approximate inference by belief propagation on the ALARM network:
311 cd path_to_libdai/matlab
312 [psi] = dai_readfg ('../tests/alarm.fg');
313 [logZ,q,md,qv,qf] = dai (psi, 'JTREE', '[updates=HUGIN,verbose=0]')
314 [logZ,q,md,qv,qf] = dai (psi, 'BP', '[updates=SEQMAX,tol=1e-9,maxiter=10000,logdomain=0]')
316 where "path_to_libdai" has to be replaced with the directory in which libDAI
317 was installed. For other algorithms and some default parameters, see the file
320 -------------------------------------------------------------------------------
322 Building the documentation
324 Install doxygen, graphviz and a TeX distribution and use
328 to build the documentation. If the documentation is not clear enough, feel free
329 to send me an email (or even better, to improve the documentation and send a
330 patch!). The documentation can also be browsed online at http://www.libdai.org.