KO | EN

GitLite

Minimalist Open Source Repository Search Engine

gitlite — search

#dsh #ai-agents #dsh-plugin #codex #developer-tools #claude-code #windows #cli #desktop-app #typescript #rust #ai-agent

Latent-Dirichlet-Allocation

★ 27 Open GitHub ↗

new version

// repository documentation

Was this content helpful?

★ 0 (0 ratings)

Latent Dirichlet Allocation 
================================

A library for the LDA topic modelling algorithm in Python and C.

Usage
-----

The best way to use `liblda` is through the command line:

./run.py --docs docs.txt --numT 40 --vocab vocab.txt --seed 3 --iter 400 --alpha 0.1 --beta 0.01 --save_probs --print_topics 10

where: 
  * `docs.txt` contains one document per line,
  * `vocab.txt` contains the vocabulary (one word per line)
  * `--save_probs` indicates that you want to output the probs phi and theta

Installation
------------

Place the directory `liblda` somewhere in your Python path.

Features
--------

We have implemented the Gibbs sampling approach which is fairly efficient when done in C.
All the rest of the functionality is done in Python so it is very hackable.

Requirements
------------
  * `numpy` (for arrays)
  * `scipy` (for weave)

Project status
--------------

The code base works, but is a bit of a mess right now.
A rewrite has begun -- in cython.

Author
--------

Ivan Savov, first dot last at gmail

Latent-Dirichlet-Allocation

Explore Similar Repositories

Was this content helpful?

Recent Feedback

Download README