- First & second-order sequential CRFs.
- Training algorithms: L-BFGS (from Taku Kudo), stochastic gradient ascent and Collin's voted perceptron.
- Handle missing training labels.
- Decoding: Viterbi (forward/backward) and Pearl's max-product.
- Report label-wise and segment-wise statistics.
Thursday, 1 May 2008
A generic code for sequential conditional random fields
I have made available the C++ code for sequential labelling tasks such as part-of-speech tagging, noun-phrase chunking or human activity annotation. The code is domain independent and supports the following features: