A simple and fast discriminative sequence labeling toolkit ( http://wapiti.limsi.fr )
Clone or download
Jekub Release v1.5.0
Signed-off-by: Thomas Lavergne <thomas.lavergne@reveurs.org>
Latest commit 569fbe5 Dec 18, 2013


Wapiti - A linear-chain CRF tool

Copyright (c) 2009-2013  CNRS
All rights reserved.

For more detailed information see the homepage.

Wapiti is a very fast toolkit for segmenting and labeling sequences with discriminative models. It is based on maxent models, maximum entropy Markov models and linear-chain CRF and proposes various optimization and regularization methods to improve both the computational complexity and the prediction performance of standard models. Wapiti is ranked first on the sequence tagging task for more than a year on MLcomp web site.

Wapiti is developed by LIMSI-CNRS and was partially funded by ANR projects CroTaL (ANR-07-MDCO-003) and MGA (ANR-07-BLAN-0311-02).

For suggestions, comments, or patchs, you can contact me at lavergne@limsi.fr

If you use Wapiti for research purpose, please use the following citation:

    author    = {Lavergne, Thomas and Capp\'{e}, Olivier and Yvon,
    title     = {Practical Very Large Scale {CRFs}},
    booktitle = {Proceedings the 48th Annual Meeting of the Association
                 for Computational Linguistics ({ACL})},
    month     = {July},
    year      = {2010},
    location  = {Uppsala, Sweden},
    publisher = {Association for Computational Linguistics},
    pages     = {504--513},
    url       = {http://www.aclweb.org/anthology/P10-1052}