moses-smt/mosesdecoder
1
1
Fork 0
mirror of https://github.com/moses-smt/mosesdecoder.git synced 2024-12-28 14:32:38 +03:00
Code Issues Packages Projects Releases Wiki Activity

gzip output phrase-based ttable by default

Browse Source
This commit is contained in:
Ondrej Bojar 2012-02-12 00:09:25 +01:00
parent e85662045d
commit 1363583c4f
1 changed files with 11 additions and 1 deletions
Show Stats Download Patch File Download Diff File Expand all files Collapse all files

12
scripts/training/filter-model-given-input.pl
View File

@ -37,8 +37,10 @@ my $ZCAT = "gzip -cd";
my $opt_hierarchical = 0;
my $binarizer = undef;
my $opt_min_non_initial_rule_count = undef;
my $opt_gzip = 1; # gzip output files (so far only phrase-based ttable until someone tests remaining models and formats)
GetOptions(
"gzip!" => \$opt_gzip,
"Hierarchical" => \$opt_hierarchical,
"Binarizer=s" => \$binarizer,
"MinNonInitialRuleCount=i" => \$opt_min_non_initial_rule_count
@ -118,6 +120,7 @@ while(<INI>) {
elsif ($binarizer && $phrase_table_impl == 0) {
print INI_OUT "1 $source_factor $t $w $new_name\n";
} else {
$new_name .= ".gz" if $opt_gzip;
print INI_OUT "$phrase_table_impl $source_factor $t $w $new_name\n";
}
push @TABLE_NEW_NAME,$new_name;
@ -223,7 +226,14 @@ for(my $i=0;$i<=$#TABLE;$i++) {
$openstring = "< $file";
}
open(FILE_OUT,">$new_file") or die "Can't write $new_file";
my $new_openstring;
if ($new_file =~ /\.gz$/) {
$new_openstring = "| gzip -c > $new_file";
} else {
$new_openstring = ">$new_file";
}
open(FILE_OUT,$new_openstring) or die "Can't write to $new_openstring";
if ($opt_hierarchical) {
my $tmp_input = $TMP_INPUT_FILENAME{$factors};