Blame view
Scripts/utils/reverse_lm_test.sh
4.05 KB
ec85f8892
first commit
|
1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60 61 62 63 64 65 66 67 68 69 70 71 72 73 74 75 76 77 78 79 80 81 82 83 84 85 86 87 88 89 90 91 |
#!/bin/bash
# Copyright 2012 Brno University of Technology (Author: Mirko Hannemann)
# Apache 2.0
# configuration section
utterances=4
maxlen=30
nbest=10
# end config section
echo "$0 $@" # Print the command line for logging
[ -f ./path.sh ] && . ./path.sh; # source the path.
. parse_options.sh || exit 1;
if [ $# != 2 ]; then
echo "Usage: utils/reverse_lm_test.sh [options] <fwd-lm-dir> <bwd-lm-dir>"
echo "example: utils/reverse_lm_test.sh data/lang_test_tgpr_5k data/lang_test_tgpr_5k.reverse"
echo "options:"
echo " --utterances <int> number of random test utterances"
echo " --maxlen <int> max number of arcs (words) in utterance"
echo " --nbest <int> compare n best paths"
exit 1;
fi
test_fwd=$1
test_bwd=$2
nb=`echo $nbest | awk '{print $1-1;}'`
# For each language model the corresponding FST in lang_test_* directory.
echo "compare LM scores using "$test_fwd/G.fst" and "$test_bwd/G.fst
for utt in `seq 1 $utterances`
do
# generate random sentence with forward language model
len=1000 # big number
while [ $len -gt $maxlen ]
do
fstrandgen --npath=1 $test_fwd/G.fst | fstprint --acceptor --isymbols=$test_fwd/words.txt --osymbols=$test_fwd/words.txt > sent$utt
len=`cat sent$utt | wc -l`
done
cat sent$utt | awk '(NF>1){if ($3!="#0") {a[length(a)+1]=$3;}} END{printf "utterance:"; for(i=1;i<=length(a);i++) {printf " %s",a[i];} printf "
";}'
# get n best paths with forward language model
cat sent$utt | awk '(NF>1){if ($3!="#0") {a[length(a)+1]=$3;}} END{for(i=1;i<=length(a);i++) {print i-1,i,a[i];} print length(a);}' > sent$utt.forward
fstcompile --acceptor --isymbols=$test_fwd/words.txt --osymbols=$test_fwd/words.txt sent$utt.forward > sent$utt.forward.fst
fstcompose $test_fwd/G.fst sent$utt.forward.fst > sent$utt.composed.forward.fst
fstshortestpath --nshortest=$nbest sent$utt.composed.forward.fst | fstprint > sent$utt.composed.forward.n
rm sent$utt.forward.scores 2>/dev/null
for n in `seq 0 $nb`
do
# select path with rank n
cat sent$utt.composed.forward.n | awk '(NR>'$n' || $1!="0"){print;}' | fstcompile | fstconnect > sent$utt.composed.forward.$n.fst
fstprint sent$utt.composed.forward.$n.fst > sent$utt.composed.forward.$n
# compute shortest distance to final states
fstshortestdistance sent$utt.composed.forward.$n.fst | \
awk -v list=sent$utt.composed.forward.$n 'BEGIN{mincost=1E5; while (getline < list > 0){if (NF==2) final[$1]=$2; if (NF==1) final[$1]=0.00001;}} \
{ if (final[$1]) { cost=$2+final[$1]; if (cost<mincost) {mincost=cost;} };} END {print mincost;}' \
>> sent$utt.forward.scores
done
# get n best paths with reverse language model
cat sent$utt | awk '(NF>1){if ($3!="#0") {a[length(a)+1]=$3;}} END{for(i=1;i<=length(a);i++) {print i-1,i,a[length(a)-i+1];} print length(a);}' > sent$utt.reverse
fstcompile --acceptor --isymbols=$test_fwd/words.txt --osymbols=$test_fwd/words.txt sent$utt.reverse > sent$utt.reverse.fst
fstcompose $test_bwd/G.fst sent$utt.reverse.fst > sent$utt.composed.reverse.fst
fstshortestpath --nshortest=$nbest sent$utt.composed.reverse.fst | fstprint > sent$utt.composed.reverse.n
rm sent$utt.reverse.scores 2>/dev/null
for n in `seq 0 $nb`
do
# select path with rank n
cat sent$utt.composed.reverse.n | awk '(NR>'$n' || $1!="0"){print;}' | fstcompile | fstconnect > sent$utt.composed.reverse.$n.fst
fstprint sent$utt.composed.reverse.$n.fst > sent$utt.composed.reverse.$n
# compute shortest distance to final states
fstshortestdistance sent$utt.composed.reverse.$n.fst | \
awk -v list=sent$utt.composed.reverse.$n 'BEGIN{mincost=1E5; while (getline < list > 0){if (NF==2) final[$1]=$2; if (NF==1) final[$1]=0.00001;}} \
{ if (final[$1]) { cost=$2+final[$1]; if (cost<mincost) {mincost=cost;} };} END {print mincost;}' \
>> sent$utt.reverse.scores
done
# present results
paste sent$utt.forward.scores sent$utt.reverse.scores | \
awk '{diff=$1-$2; if ( (diff<0?-diff:diff) > 0.001 ) print NR,$1,$2,"!!!"; else print NR,$1,$2;}'
# clean up
rm sent$utt
rm sent$utt.*
done
|