mloss.org CRFsuitehttp://mloss.orgUpdates and additions to CRFsuiteenWed, 18 Mar 2009 15:19:02 -0000CRFsuite 0.8http://mloss.org/software/view/179/<html><p>CRFSuite is an implementation of Conditional Random Fields (CRFs) for labeling sequential data. Among the various implementations of CRFs, this software provides following features. </p> <ul> <li> Speed-oriented implementation written in pure C. </li> <li> Very fast parameter estimation using Stochastic Gradient Descent (SGD). </li> <li> Fast parameter estimation using Limited-memory BFGS (L-BFGS). </li> <li> L1 regularization (Laplacian prior) using Orthant-Wise Limited-memory Quasi-Newton (OWL-QN) method. </li> <li> L2 regularization (Gaussian prior). </li> <li> Performance evaluation on training. </li> <li> Simple data I/O format. </li> <li> An efficient file format for storing/accessing CRF models. </li> </ul></html>Naoaki OkazakiWed, 18 Mar 2009 15:19:02 -0000http://mloss.org/software/rss/comments/179http://mloss.org/software/view/179/structure learningcrfstochastic gradient descentconditional random fieldssequential labelingsgd