Age | Commit message (Collapse) | Author | |
---|---|---|---|
2015-03-03 | migration to cmake | Chris Dyer | |
2014-10-09 | make tree terminals available to feature functions | Chris Dyer | |
2014-09-22 | added missing c++11 headers (could not compile with gcc-4.8) | Wilker Aziz | |
2014-08-02 | finally | Patrick Simianer | |
2014-08-02 | fix | Patrick Simianer | |
2014-07-29 | make PassThrough1..N optional | Patrick Simianer | |
2014-04-07 | clean up dead TRule code | Chris Dyer | |
2014-02-02 | load multiple grammars | Chris Dyer | |
2013-12-26 | add support for epsilons in input lattice | Chris Dyer | |
2013-11-10 | guard against direct includes of tr1 | Chris Dyer | |
2013-05-06 | add passthrough length features | Chris Dyer | |
2012-10-25 | move some grammar stuff around | Chris Dyer | |
2012-07-20 | fix | Chris Dyer | |
2012-07-20 | enable programmatic setting of grammar objects, cleanup of grammar handling | Chris Dyer | |
2012-05-27 | clean up | Chris Dyer | |
2012-04-29 | reverted changes in upstream | Patrick Simianer | |
2011-09-23 | big update: working iterating, pretty output, test scripts and more | Patrick Simianer | |
2011-09-23 | hacking in weights setting, getting | Patrick Simianer | |
2011-09-23 | added dtrain, implemented psg setting in decoder/translator, implemented ↵ | Patrick Simianer | |
psmetrics, ngram statistics, main carcass | |||
2011-04-28 | removing logging | Chris Dyer | |
2011-04-22 | make compatible with FastSparseVector | Chris Dyer | |
2011-02-10 | multipass decoding | Chris Dyer | |
2010-12-12 | facility for adding programmitically generated grammars | Chris Dyer | |
2010-10-05 | add PrepareForInput to ff interface, make sgml markup available to feature ↵ | redpony | |
functions git-svn-id: https://ws10smt.googlecode.com/svn/trunk@669 ec762483-ff6d-05da-a07a-a48fb63a330f | |||
2010-09-21 | fix for lattice cost bug | redpony | |
git-svn-id: https://ws10smt.googlecode.com/svn/trunk@653 ec762483-ff6d-05da-a07a-a48fb63a330f | |||
2010-09-20 | support turning off verbose logging | redpony | |
git-svn-id: https://ws10smt.googlecode.com/svn/trunk@652 ec762483-ff6d-05da-a07a-a48fb63a330f | |||
2010-07-24 | FSA: simpler Scan1 ScanT1 methods, otherewise also expose edge to full Scan | graehl | |
git-svn-id: https://ws10smt.googlecode.com/svn/trunk@399 ec762483-ff6d-05da-a07a-a48fb63a330f | |||
2010-07-15 | oracle bleu refactor | graehl | |
git-svn-id: https://ws10smt.googlecode.com/svn/trunk@259 ec762483-ff6d-05da-a07a-a48fb63a330f | |||
2010-07-13 | cdec now supports coarse-to-fine decoding (for SCFG only). | adam.d.lopez | |
CTF has several options: -coarse_to_fine_beam_prune=<double> (required to activate CTF) assign an alpha parameter for pruning the coarse foreast -ctf_beam_widen=<double> (optional, defaults to 2.0): ratio to widen coarse pruning beam if fine parse fails. -ctf_num_widenings=<int> (optional, defaults to 2): number of times to widen coarse beam before defaulting to exhaustive source parsing -ctf_no_exhaustive (optional) do not attempt exhaustive parse if CTF fails to find a parse. Additionally, script extools/coarsen_grammar.pl will create a coarse-to-fine grammar (for X?? categories *only*). cdec will read CTF grammars in a format identical to the original, in which refinements of a rule immediately follow the coarse projection, preceded by an additional whitespace character. Not fully tested, but should be backwards compatible. Also not yet integrated into pipelines, but should work on the command line. git-svn-id: https://ws10smt.googlecode.com/svn/trunk@231 ec762483-ff6d-05da-a07a-a48fb63a330f | |||
2010-07-05 | TODO (mostly low priority) notes | graehl | |
git-svn-id: https://ws10smt.googlecode.com/svn/trunk@128 ec762483-ff6d-05da-a07a-a48fb63a330f | |||
2010-06-22 | initial checkin | redpony | |
git-svn-id: https://ws10smt.googlecode.com/svn/trunk@2 ec762483-ff6d-05da-a07a-a48fb63a330f |