summaryrefslogtreecommitdiff
path: root/decoder
AgeCommit message (Collapse)Author
2010-01-29word aligner checkinChris Dyer
2010-01-27add binary model2-like featuresChris Dyer
2010-01-25fix constChris Dyer
2010-01-24more autoconf fixes- use version of boost m4 macros which are much, much, ↵Chris Dyer
much better
2010-01-24Support building without gtestChris Dyer
Now the only dependence is boost, which most modern linux distros have.
2009-12-26add inside algorithm pass in pruning (prevents parent nodes that are ↵Chris Dyer
underivable from hanging around)
2009-12-26minor fix for fast intersection codeChris Dyer
2009-12-26increase intersection speed by a couple orders of magnitude for linear chain ↵Chris Dyer
graphs
2009-12-19few fixesChris Dyer
2009-12-19cool new alignment featureChris Dyer
2009-12-19fix rounding error failureChris Dyer
2009-12-18add symmetrization heuristics to atools, add null word configurationChris Dyer
2009-12-18add support for freezing the feature set to a user-specified list, even if ↵Chris Dyer
feature detectors create additional features
2009-12-17make node order explicit in graphChris Dyer
2009-12-17added non-pruning intersection and a CRF taggerChris Dyer
- the linear-chain tagger is more of a proof of concept than a real tagger-- the context-free assumptions made in a number of places mean that the algorithms used may not be as efficient as they could be, but the model is as powerful as any CRF - it would be easy to add latent variables or semi-CRF support (or both!) - i've added a couple basic features that are often used for POS tagging - non-pruning intersection is useful for lexical word alignment models and the tagger - a sample POS tagger model will be committed later
2009-12-14few small fixes of alignment tools, add new orthographic similarity feature ↵Chris Dyer
for word aligner, final naming of directories, libraries in cdec