Topic modeling with latent Dirichlet allocation. lda aims for simplicity.

lda implements latent Dirichlet allocation (LDA) using collapsed Gibbs sampling. LDA is described in Blei et al. (2003) and Pritchard et al. (2000).

pip install lda

lda.LDA implements latent Dirichlet allocation (LDA). The interface follows conventions found in scikit-learn.

>>> import numpy as np
>>> import lda
>>> X = np.array([[1,1], [2, 1], [3, 1], [4, 1], [5, 8], [6, 1]])
>>> model = lda.LDA(n_topics=2, n_iter, random_state=1)
>>> doc_topic = model.fit_transform(X)  # estimate of document-topic distributions
>>> model.components_  # estimate of topic-word distributions; model.doc_topic_ is an alias

Python 2.7 or Python 3.3+ is required. The following packages are required

• numpy
• scipy
• pbr

lda aims for simplicity. (It happens to be fast, as essential parts are written in C via Cython_.) If you are working with a very large corpus you may wish to use more sophisticated topic models such as those implemented in hca and MALLET. hca is written in C and MALLET_ is written in Java. Unlike lda, hca can use more than one processor at a time.

• Documentation: http://pythonhosted.org/lda
• Source code: https://github.com/ariddell/lda/
• Issue tracker: https://github.com/ariddell/lda/issues

lda is licensed under Version 2.0 of the Mozilla Public License.