GitHub - vats-div/JunctionTreeUGMS: Junction Tree Framework for Undirected Graphical Model Selection

vats-div / JunctionTreeUGMS Public

Notifications You must be signed in to change notification settings
Fork 2
Star 2

Junction Tree Framework for Undirected Graphical Model Selection

Notifications

Name		Name	Last commit message	Last commit date
Latest commit History 1 Commit
KPMtools		KPMtools
L1General		L1General
L1GeneralGroup		L1GeneralGroup
LogdetPPA-0		LogdetPPA-0
MuG		MuG
RealData		RealData
SyntheticExamples		SyntheticExamples
UGMSAlgorithms		UGMSAlgorithms
graphFunctions		graphFunctions
minConf		minConf
myFunctions		myFunctions
ExampleScript.m		ExampleScript.m
README.txt		README.txt
configurepathMaC.m		configurepathMaC.m
configurepathWin.m		configurepathWin.m

Repository files navigation

We provide Matlab code for the algorithms in the paper:

D. Vats and R. D. Nowak
A Junction Tree Framework for Undirected Graphical Model Selection

USAGE
-----
To use the scripts, you will first need to include all the directories in the Matlab path.
Note that many functions included are borrowed from other sources (mainly http://code.google.com/p/pmtk3/)

If you are on a Windows machine, run configurepathWin.

If you are on a MaC/Linux/Unix machine, run configurepathMac.

See ExampleScript.m for examples of using the various functions in the toolbox.

See the folder RealData/ for the results on real data given in the paper.

See the folder SyntheticExamples for scripts used to generate Tables 2-5 in the paper.  If you would like the data we used for the synthetic graphs, email at [email protected]. 

UGMS ALGORITHMS
---------------

In the folder UGMSAlgorithms, we have provided implementations of three UGMS algorithms:

- graphical Lasso (UGMS_GLasso.m)
- neighborhood selection using adaptive Lasso (UGMS_NLasso.m)
- PC-Algorithm (UGMS_PC.m)

See the individual scripts or ExampleScript.m on how to use these functions.  

CONTACT
------

Divyanshu Vats
[email protected] or [email protected]