wbuntine has posted 4 projects.


Logo hca 0.6

by wbuntine - August 6, 2014, 14:24:57 CET [ Project Homepage BibTeX BibTeX for corresponding Paper Download ] 3387 views, 622 downloads, 3 subscriptions

About: Multi-core non-parametric and bursty topic models (HDP-LDA, DCMLDA, and other variants of LDA) implemented in C using efficient Gibbs sampling, with hyperparameter sampling and other flexible controls.

Changes:

Modified command line -A and -B formats. Overhaul of diagnostics. Described changes in manual. Bug fixes: multi-core crashing when huge number of topics; -B when using number and fitting beta, beta sampling wasn't working; both now fixed.


Logo DCABags 0.7

by wbuntine - June 5, 2014, 05:34:44 CET [ Project Homepage BibTeX Download ] 1960 views, 454 downloads, 4 subscriptions

About: Document/Text preprocessing for topic models: suite of Perl scripts for preprocessing text collections to create dictionaries and bag/list files for use by topic modelling software.

Changes:

Moved distribution and code across to GitHub. Changed "ldac" format to have 0 offset for word indices. Added "document frequency" (df) filtering on selection of tokens for linkTables. Playing with linkParse but its still unuseable generally.


Logo libstb 1.8

by wbuntine - April 24, 2014, 09:02:17 CET [ Project Homepage BibTeX BibTeX for corresponding Paper Download ] 4397 views, 850 downloads, 1 subscription

About: Generalised Stirling Numbers for Pitman-Yor Processes: this library provides ways of computing generalised 2nd-order Stirling numbers for Pitman-Yor and Dirichlet processes. Included is a tester and parameter optimiser. This accompanies Buntine and Hutter's article: http://arxiv.org/abs/1007.0296, and a series of papers by Buntine and students at NICTA and ANU.

Changes:

Moved repository to GitHub, and added thread support to use the main table lookups in multi-threaded code.


Logo Generalised Stirling Numbers libstb 1.0 1.4

by wbuntine - September 28, 2012, 13:49:57 CET [ Project Homepage BibTeX BibTeX for corresponding Paper Download ] 4638 views, 871 downloads, 1 subscription

About: THIS VERSION DISCONTINUED, see "http://mloss.org/software/view/424/". This library provides ways of computing generalised 2nd-order Stirling numbers for Pitman-Yor and Dirichlet processes. Included is a tester and parameter optimiser. This accompanies Buntine and Hutter's article: http://arxiv.org/abs/1007.0296

Changes:

See the alternative MLOSS entry "libstb". Updated to 1.4!