Skip to content

guaibaoer/topic

 
 

Folders and files

NameName
Last commit message
Last commit date

Latest commit

 

History

14 Commits
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 

Repository files navigation

Sparse Gibbs sampler for LDA

This is a minimalistic C++11 implementation of Sparse sampling for LDA.

This package depends on glog, gflags, and gperftools(optional). To build,

cd some_directory
git clone https://github.com/xunzheng/third_party
cd third_party
./install.sh

Third party libraries will be installed at some_directory/third_party/.

Now we can build the project:

git clone https://github.com/xunzheng/topic
cd topic
ln -s some_directory/third_party third_party
make

Toy dataset 20news.train is included in the exp/ directory. Try

cd exp
./run.sh

to get a sample run. Since LDA is an unsupervised model, label information in LIBSVM format is ignored.

To see all the available flags, run

./gibbs

without any flags.

Reference

Limin Yao, David Mimno, and Andrew McCallum. Efficient Methods for Topic Model Inference on Streaming Document Collections. In International Conference on Knowledge Discovery and Data mining (SIGKDD), 2009.

About

Sparse sampler for LDA

Resources

Stars

Watchers

Forks

Releases

No releases published

Packages

No packages published

Languages

  • C++ 95.0%
  • Makefile 2.7%
  • Python 2.1%
  • Shell 0.2%