From e26434979adc33bd949566ba7bf02dff64e80a3e Mon Sep 17 00:00:00 2001 From: Chris Dyer Date: Tue, 2 Oct 2012 00:19:43 -0400 Subject: cdec cleanup, remove bayesian stuff, parsing stuff --- gi/pipeline/scripts/xfeats.pl | 39 --------------------------------------- 1 file changed, 39 deletions(-) delete mode 100755 gi/pipeline/scripts/xfeats.pl (limited to 'gi/pipeline/scripts/xfeats.pl') diff --git a/gi/pipeline/scripts/xfeats.pl b/gi/pipeline/scripts/xfeats.pl deleted file mode 100755 index dc578513..00000000 --- a/gi/pipeline/scripts/xfeats.pl +++ /dev/null @@ -1,39 +0,0 @@ -#!/usr/bin/perl -w -use strict; - -die "Usage: $0 x-grammar.scfg[.gz] < cat-grammar.scfg\n" unless scalar @ARGV > 0; - -my $xgrammar = shift @ARGV; -die "Can't find $xgrammar" unless -f $xgrammar; -my $fh; -if ($xgrammar =~ /\.gz$/) { - open $fh, "gunzip -c $xgrammar|" or die "Can't fork: $!"; -} else { - open $fh, "<$xgrammar" or die "Can't read $xgrammar: $!"; -} -print STDERR "Reading X-feats from $xgrammar...\n"; -my %dict; -while(<$fh>) { - chomp; - my ($lhs, $f, $e, $feats) = split / \|\|\| /; - my $xfeats; - my $cc = 0; - my @xfeats = (); - while ($feats =~ /(EGivenF|FGivenE|LogRuleCount|LogECount|LogFCount|SingletonRule|SingletonE|SingletonF)=([^ ]+)( |$)/og) { - push @xfeats, "X_$1=$2"; - } - #print "$lhs ||| $f ||| $e ||| @xfeats\n"; - $dict{"$lhs ||| $f ||| $e"} = "@xfeats"; -} -close $fh; - -print STDERR "Add features...\n"; -while(<>) { - chomp; - my ($lhs, $f, $e) = split / \|\|\| /; - $f=~ s/\[[^]]+,([12])\]/\[X,$1\]/g; - my $xfeats = $dict{"[X] ||| $f ||| $e"}; - die "Can't find x features for: $_\n" unless $xfeats; - print "$_ $xfeats\n"; -} - -- cgit v1.2.3