CRF--, a MATLAB version CRF toolkit, which can use train/test data from CRF++.
CRF-- is integrated with both IIS (algorithm S in the original paper:Conditional Random Fields (CRFs) and LBFGS.
* Note that CRF-- is not complete yet, I will release it ASAP.
[19-apr-2011 14:58:25]
Integrated with very fast L-BFGS(fminlbfgs_version2c) optimizer. (mod config_file to select optimizer)
* fminlbfgs_version2c: Copyright (c) 2009, Dirk-Jan Kroon
[25-Mar-2011 21:40:45]
Fast matrix M computation
[19-Mar-2011 11:52:14]
Fast expectation of feat' func' w.r.t model computation
Demo script added
Viterbi decoding algorithm
Viterbi step size algorithm
Visualization(not functional)
Please note my website or my email address if you want to share these tools with the others.
If you have any good ideas please contact me through Email: szhiliang@gmail.com