summaryrefslogtreecommitdiff
path: root/gi
AgeCommit message (Collapse)Author
2010-07-23Fixed bug when patching the source language with tags and preserving phrase ↵trevor.cohn
tokens. git-svn-id: https://ws10smt.googlecode.com/svn/trunk@378 ec762483-ff6d-05da-a07a-a48fb63a330f
2010-07-23Pipeline code for running with mixing tokens and tags in the clustering.trevor.cohn
git-svn-id: https://ws10smt.googlecode.com/svn/trunk@377 ec762483-ff6d-05da-a07a-a48fb63a330f
2010-07-23Changed the initialisation of the sampler, hopefully this will work better.philblunsom
git-svn-id: https://ws10smt.googlecode.com/svn/trunk@376 ec762483-ff6d-05da-a07a-a48fb63a330f
2010-07-22variational bayes inferencedesaicwtf
git-svn-id: https://ws10smt.googlecode.com/svn/trunk@372 ec762483-ff6d-05da-a07a-a48fb63a330f
2010-07-22forgottenredpony
git-svn-id: https://ws10smt.googlecode.com/svn/trunk@371 ec762483-ff6d-05da-a07a-a48fb63a330f
2010-07-22fixed typotrevor.cohn
git-svn-id: https://ws10smt.googlecode.com/svn/trunk@370 ec762483-ff6d-05da-a07a-a48fb63a330f
2010-07-22Fixed filename clash error when running with tagged corpus and source and ↵trevor.cohn
target laguagages git-svn-id: https://ws10smt.googlecode.com/svn/trunk@369 ec762483-ff6d-05da-a07a-a48fb63a330f
2010-07-22add additional filtering stepredpony
git-svn-id: https://ws10smt.googlecode.com/svn/trunk@368 ec762483-ff6d-05da-a07a-a48fb63a330f
2010-07-22Added option to apply tags to source-sidetrevor.cohn
git-svn-id: https://ws10smt.googlecode.com/svn/trunk@367 ec762483-ff6d-05da-a07a-a48fb63a330f
2010-07-21Posterior outputtrevor.cohn
git-svn-id: https://ws10smt.googlecode.com/svn/trunk@366 ec762483-ff6d-05da-a07a-a48fb63a330f
2010-07-21Fixes to PR command line.trevor.cohn
Added bilingual agreement model processing to pipeline. git-svn-id: https://ws10smt.googlecode.com/svn/trunk@365 ec762483-ff6d-05da-a07a-a48fb63a330f
2010-07-21Fixing errors.olivia.buzek
git-svn-id: https://ws10smt.googlecode.com/svn/trunk@361 ec762483-ff6d-05da-a07a-a48fb63a330f
2010-07-21Debugging backoff.olivia.buzek
git-svn-id: https://ws10smt.googlecode.com/svn/trunk@359 ec762483-ff6d-05da-a07a-a48fb63a330f
2010-07-21Agreement model flagtrevor.cohn
git-svn-id: https://ws10smt.googlecode.com/svn/trunk@358 ec762483-ff6d-05da-a07a-a48fb63a330f
2010-07-21Little bug fix to EM clusteringtrevor.cohn
git-svn-id: https://ws10smt.googlecode.com/svn/trunk@357 ec762483-ff6d-05da-a07a-a48fb63a330f
2010-07-21previously committed wrong versiondesaicwtf
git-svn-id: https://ws10smt.googlecode.com/svn/trunk@355 ec762483-ff6d-05da-a07a-a48fb63a330f
2010-07-21 corpus reads optional tags from data, EM trains with those tags, fix a bug ↵desaicwtf
in PhraseCluster where phrase priors are not learned git-svn-id: https://ws10smt.googlecode.com/svn/trunk@354 ec762483-ff6d-05da-a07a-a48fb63a330f
2010-07-20minor fix in span agreement vizbothameister
git-svn-id: https://ws10smt.googlecode.com/svn/trunk@349 ec762483-ff6d-05da-a07a-a48fb63a330f
2010-07-20removed filetrevor.cohn
git-svn-id: https://ws10smt.googlecode.com/svn/trunk@344 ec762483-ff6d-05da-a07a-a48fb63a330f
2010-07-20fixed typotrevor.cohn
git-svn-id: https://ws10smt.googlecode.com/svn/trunk@343 ec762483-ff6d-05da-a07a-a48fb63a330f
2010-07-20Added backoff pipe.olivia.buzek
git-svn-id: https://ws10smt.googlecode.com/svn/trunk@340 ec762483-ff6d-05da-a07a-a48fb63a330f
2010-07-20Allow fractional counts.trevor.cohn
git-svn-id: https://ws10smt.googlecode.com/svn/trunk@337 ec762483-ff6d-05da-a07a-a48fb63a330f
2010-07-20Cleaned up scriptstrevor.cohn
git-svn-id: https://ws10smt.googlecode.com/svn/trunk@336 ec762483-ff6d-05da-a07a-a48fb63a330f
2010-07-20Fixed bug in mpi output.philblunsom
git-svn-id: https://ws10smt.googlecode.com/svn/trunk@335 ec762483-ff6d-05da-a07a-a48fb63a330f
2010-07-20git-svn-id: https://ws10smt.googlecode.com/svn/trunk@334 ↵desaicwtf
ec762483-ff6d-05da-a07a-a48fb63a330f
2010-07-20modified trainer for agreement of languagesdesaicwtf
git-svn-id: https://ws10smt.googlecode.com/svn/trunk@333 ec762483-ff6d-05da-a07a-a48fb63a330f
2010-07-19A slightly more general version.philblunsom
git-svn-id: https://ws10smt.googlecode.com/svn/trunk@328 ec762483-ff6d-05da-a07a-a48fb63a330f
2010-07-19Tool to pull out separate language data from context.txt.gztrevor.cohn
git-svn-id: https://ws10smt.googlecode.com/svn/trunk@326 ec762483-ff6d-05da-a07a-a48fb63a330f
2010-07-19Fixed command linetrevor.cohn
git-svn-id: https://ws10smt.googlecode.com/svn/trunk@325 ec762483-ff6d-05da-a07a-a48fb63a330f
2010-07-19Reversed out broken thresholdingtrevor.cohn
git-svn-id: https://ws10smt.googlecode.com/svn/trunk@324 ec762483-ff6d-05da-a07a-a48fb63a330f
2010-07-19MPI stuff: hierarchical topics should work.philblunsom
git-svn-id: https://ws10smt.googlecode.com/svn/trunk@321 ec762483-ff6d-05da-a07a-a48fb63a330f
2010-07-19Vaguely working distributed implementation. Hierarchical topics doesn't yet ↵philblunsom
work correctly. git-svn-id: https://ws10smt.googlecode.com/svn/trunk@317 ec762483-ff6d-05da-a07a-a48fb63a330f
2010-07-18??trevor.cohn
git-svn-id: https://ws10smt.googlecode.com/svn/trunk@312 ec762483-ff6d-05da-a07a-a48fb63a330f
2010-07-18Changed to UTF8trevor.cohn
git-svn-id: https://ws10smt.googlecode.com/svn/trunk@311 ec762483-ff6d-05da-a07a-a48fb63a330f
2010-07-17more support for other clustersredpony
git-svn-id: https://ws10smt.googlecode.com/svn/trunk@307 ec762483-ff6d-05da-a07a-a48fb63a330f
2010-07-17support for running on different clustersredpony
git-svn-id: https://ws10smt.googlecode.com/svn/trunk@306 ec762483-ff6d-05da-a07a-a48fb63a330f
2010-07-17Extra optionstrevor.cohn
git-svn-id: https://ws10smt.googlecode.com/svn/trunk@305 ec762483-ff6d-05da-a07a-a48fb63a330f
2010-07-16Fixed PR command linetrevor.cohn
git-svn-id: https://ws10smt.googlecode.com/svn/trunk@303 ec762483-ff6d-05da-a07a-a48fb63a330f
2010-07-16agreement between source and target sidedesaicwtf
git-svn-id: https://ws10smt.googlecode.com/svn/trunk@300 ec762483-ff6d-05da-a07a-a48fb63a330f
2010-07-16Added picturestrevor.cohn
git-svn-id: https://ws10smt.googlecode.com/svn/trunk@299 ec762483-ff6d-05da-a07a-a48fb63a330f
2010-07-16Added various flags to filter out low count events (words, edges).trevor.cohn
git-svn-id: https://ws10smt.googlecode.com/svn/trunk@298 ec762483-ff6d-05da-a07a-a48fb63a330f
2010-07-16addlinh.kitty
git-svn-id: https://ws10smt.googlecode.com/svn/trunk@291 ec762483-ff6d-05da-a07a-a48fb63a330f
2010-07-16addlinh.kitty
git-svn-id: https://ws10smt.googlecode.com/svn/trunk@286 ec762483-ff6d-05da-a07a-a48fb63a330f
2010-07-16git-svn-id: https://ws10smt.googlecode.com/svn/trunk@285 ↵philblunsom
ec762483-ff6d-05da-a07a-a48fb63a330f
2010-07-16working on mpi implementation.philblunsom@gmail.com
git-svn-id: https://ws10smt.googlecode.com/svn/trunk@283 ec762483-ff6d-05da-a07a-a48fb63a330f
2010-07-15Option to run on single word phrases before moving to larger ones.trevor.cohn
git-svn-id: https://ws10smt.googlecode.com/svn/trunk@272 ec762483-ff6d-05da-a07a-a48fb63a330f
2010-07-15Unified backoff_grammar and hier_cat.olivia.buzek
git-svn-id: https://ws10smt.googlecode.com/svn/trunk@270 ec762483-ff6d-05da-a07a-a48fb63a330f
2010-07-15generalised parsing of 'features' (in clustering output) during span labellingbothameister
git-svn-id: https://ws10smt.googlecode.com/svn/trunk@269 ec762483-ff6d-05da-a07a-a48fb63a330f
2010-07-15rolled back unintended commit on *evaluation* pipelinebothameister
git-svn-id: https://ws10smt.googlecode.com/svn/trunk@268 ec762483-ff6d-05da-a07a-a48fb63a330f
2010-07-15updated pipeline with --use_default_cat to handle unlabelled spans (which ↵bothameister
default to 'X'); added a shortcut so that pipeline can return the directory name where a labelled corpus ends up - I find it helpful for organizing experiments git-svn-id: https://ws10smt.googlecode.com/svn/trunk@267 ec762483-ff6d-05da-a07a-a48fb63a330f