/clda

Compound Latent Dirichlet Allocation Model

Primary LanguageC++GNU General Public License v3.0GPL-3.0

Algorithms for the Compound Latent Dirichlet Allocation Model

This R package, clda, implements the following Markov chain Monte Carlo (MCMC) and variational methods for the compound latent Dirichlet allocation (cLDA) model.

  • Auxiliary variable update within Gibbs sampler (AGS)
  • Metropolis adjusted Langevin algorithm within Gibbs sampler (MGS)
  • Variational Expectation Maximization (VEM)

For package documentation run

help("clda")

in an R console. All major functions and datasets are documented and linked to the package index.

To see all demo R scripts available in this package, run

demo(package="clda")

in an R console. These demo scripts require commandline arguments for execution. Please see the documentation provided in each script before execution.

Authors

  • Clint P. George (Please contact for questions and comments)
  • Wei Xia
  • George Michailidis

Dependencies

This package uses the following R packages, which are already included in this R package.

  • Rcpp
  • RcppArmadillo based on the Armadillo C++ package
  • lattice

Installation Guide

  • Download the package source from Git Download Link
  • Unzip the dowloaded file and rename the folder clda-master to clda
  • To install clda run R CMD INSTALL clda on the commandline
  • To uninstall clda run R CMD REMOVE clda on the commandline