added filler rules for MosesCompoundSplitter

stanfordnlp / phrasal

A large-scale statistical machine translation system written in Java.

http://nlp.stanford.edu/

GNU General Public License v3.0

208 stars 88 forks source link

Closed sahasan closed 7 years ago

sahasan commented 7 years ago

test on De-En w/ a 12k test set shows improvement of 0.5% BLEU

this commit also fixes a unit test to make it pass on Windows

joewue commented 7 years ago

LGTM