summaryrefslogtreecommitdiff
path: root/decoder/ff.cc
diff options
context:
space:
mode:
authorgraehl <graehl@ec762483-ff6d-05da-a07a-a48fb63a330f>2010-07-23 21:02:44 +0000
committergraehl <graehl@ec762483-ff6d-05da-a07a-a48fb63a330f>2010-07-23 21:02:44 +0000
commit353ff197b082b71c623c7851e0df0ac60d770533 (patch)
tree3d55c79807f86cbe579df3a9d1aebf4464331423 /decoder/ff.cc
parent37f4f68effe3b5ab823985798e77fa64dd66088d (diff)
implicit first param to ffs: "debug". fsa final traversal set feature=0 first. set FF_FSA_DEBUG (also FSA_DEBUG).
git-svn-id: https://ws10smt.googlecode.com/svn/trunk@387 ec762483-ff6d-05da-a07a-a48fb63a330f
Diffstat (limited to 'decoder/ff.cc')
-rw-r--r--decoder/ff.cc2
1 files changed, 2 insertions, 0 deletions
diff --git a/decoder/ff.cc b/decoder/ff.cc
index b6a541e3..28d6f732 100644
--- a/decoder/ff.cc
+++ b/decoder/ff.cc
@@ -1,3 +1,5 @@
+//TODO: actually score rule_feature()==true features once only, hash keyed on rule or modify TRule directly? need to keep clear in forest which features come from models vs. rules; then rescoring could drop all the old models features at once
+
//TODO: 0 size state != rule-local feature, i.e. still may depend on source span loc/context. identify truly rule-local features so if we want they can be added to grammar rules (minor speedup)
#include <boost/lexical_cast.hpp>