index
:
cdec-dtrain-legacy
json_serial
master
net
Mirror of https://github.com/pks/cdec-dtrain-legacy.git
summary
refs
log
tree
commit
diff
log msg
author
committer
range
path:
root
/
compound-split
Age
Commit message (
Expand
)
Author
2014-02-10
transition away from checking in big data files
Chris Dyer
2014-01-18
new tuning of crf compound splitter for wmt14
Chris Dyer
2014-01-17
new de split
Chris Dyer
2012-11-14
major mert clean up, stuff for simple system demo
Chris Dyer
2012-11-06
fix
Chris Dyer
2012-11-06
fix up some frequencies to give better results on a few common words
Chris Dyer
2012-11-05
larger training data for semi-crf word segmenter
Chris Dyer
2012-04-29
reverted changes in upstream
Patrick Simianer
2011-11-03
Merge remote-tracking branch 'upstream/master'
Patrick Simianer
2011-11-03
local hacks
Patrick Simianer
2011-10-24
noun-segmenting weights for german word segmenter
Chris Dyer
2011-09-21
Updated kenlm. Includes left state support but not the cdec-side use of it. ...
Kenneth Heafield
2011-03-10
Fix broken klm file for de compounding
Kenneth Heafield
2011-03-10
remove dependency on SRILM
Chris Dyer
2011-02-16
minor casing bugfix
Chris Dyer
2011-02-16
add case preservation to compound splitter
Chris Dyer
2011-01-18
new version of klm
Chris Dyer
2011-01-17
more german fixes
Chris Dyer
2011-01-13
updated training data, retrained de seg model
Chris Dyer
2010-12-22
missing word
Chris Dyer
2010-12-22
small updates to german model
Chris Dyer
2010-12-22
fix compound splitter, new features, more training data
Chris Dyer
2010-06-22
initial checkin
redpony