mirror of
https://github.com/moses-smt/mosesdecoder.git
synced 2024-08-16 23:10:31 +03:00
devanagari fix
This commit is contained in:
parent
2affb9b624
commit
3910cd6c46
@ -121,7 +121,7 @@ sub preprocess {
|
||||
|
||||
# Sentences can start with upper-case, numnbers, or Indic characters
|
||||
my $sentence_start = "\\p{IsUpper}0-9";
|
||||
$sentence_start .= "\\p{Block: Devanagari_Extended}" if $language eq "hi";
|
||||
$sentence_start .= "\\p{Block: Devanagari}\\p{Block: Devanagari_Extended}" if $language eq "hi";
|
||||
$sentence_start .= "\\p{Block: Gujarati}" if $language eq "gu";
|
||||
|
||||
# we include danda and double danda (U+0964 and U+0965) as sentence split characters
|
||||
|
Loading…
Reference in New Issue
Block a user