s421818-mlworkshops/line_counter.sh
2020-04-03 13:19:26 +02:00

14 lines
438 B
Bash

wc -l wikiniews_results.tsv > liczba_linii.txt
cut -f3 wikiniews_results.tsv > reference.txt
cut -f2 wikiniews_results.tsv > hypothesis.txt
awk 'BEGIN{FS=OFS="\t"}{print $0,"(sp1_"NR")"}' < reference.txt > reference.trn
awk 'BEGIN{FS=OFS="\t"}{print $0,"(sp1_"NR")"}' < hypothesis.txt > hypothesis.trn
sclite -f 0 -r reference.trn trn -h hypothesis.trn trn -e utf-8 -i rm -o all stdout
head -15 reference.trn
head -15 hypothesis.trn