summaryrefslogtreecommitdiff
path: root/preprocess
diff options
context:
space:
mode:
authorPatrick Simianer <simianer@cl.uni-heidelberg.de>2013-12-11 16:56:56 +0100
committerPatrick Simianer <simianer@cl.uni-heidelberg.de>2013-12-11 16:56:56 +0100
commit941da973ffd77b5b0accb64cb4ffe7f593dcd5f8 (patch)
tree881640442c5171175945204c6626a08fda57a429 /preprocess
parent36e78054419ccf38fb3049e32120b5704de4ba85 (diff)
use moses tokenizer but without escaping of 'special' chars
Diffstat (limited to 'preprocess')
-rwxr-xr-xpreprocess2
1 files changed, 1 insertions, 1 deletions
diff --git a/preprocess b/preprocess
index 9273cfb..14eb73a 100755
--- a/preprocess
+++ b/preprocess
@@ -1,5 +1,5 @@
#!/bin/zsh
LANG=$1
-/toolbox/scripts/htmlentities 2>htmlentities.$LANG.err | /toolbox/scripts/normalize_punctuation 2>normalize-punctuation.$LANG.err | /toolbox/moses/scripts/tokenizer/tokenizer.perl -a -b -threads 1 -l $LANG 2>tokenizer.$LANG.err | /toolbox/moses/scripts/tokenizer/lowercase.perl 2>lowercase.$LANG.err
+/toolbox/scripts/htmlentities 2>htmlentities.$LANG.err | /toolbox/scripts/normalize_punctuation 2>normalize-punctuation.$LANG.err | /toolbox/moses/scripts/tokenizer/tokenizer.no-esc.perl -a -b -threads 1 -l $LANG 2>tokenizer.$LANG.err | /toolbox/moses/scripts/tokenizer/lowercase.perl 2>lowercase.$LANG.err