anasampa2's picture
Upload 151 files
ee0ec3d verified
raw
history blame
No virus
1.18 kB
#!/bin/bash
for l in a m onlym; do
mkdir -p pdtc-1.0-pdt-$l
case $l in
onlym) src=mw; l1=m;;
*) src=*$l*; l1=$l;;
esac
ls data/PDT/pml/$src/train*/*.$l1.gz | sed 's/\.[am]\.gz$//' > pdtc-1.0-pdt-$l/files_train
ls data/PDT/pml/$src/dtest*/*.$l1.gz | sed 's/\.[am]\.gz$//' > pdtc-1.0-pdt-$l/files_dev
ls data/PDT/pml/$src/etest*/*.$l1.gz | sed 's/\.[am]\.gz$//' > pdtc-1.0-pdt-$l/files_test
done
mkdir -p pdtc-1.0-pcedt-m
ls data/PCEDT/pml/wsj{0[0-9],1[0-8]}*.m.gz | sed 's/\.[am]\.gz$//' > pdtc-1.0-pcedt-m/files_train
ls data/PCEDT/pml/wsj{19,2[01]}*.m.gz | sed 's/\.[am]\.gz$//' > pdtc-1.0-pcedt-m/files_dev
ls data/PCEDT/pml/wsj2[234]*.m.gz | sed 's/\.[am]\.gz$//' > pdtc-1.0-pcedt-m/files_test
for t in Faust PDTSC; do
mkdir -p pdtc-1.0-${t,,}-m
ls data/$t/pml/*_{[0-9],[0-9][0-9]}[^67].*m.gz | sed 's/\.[am]\.gz$//' > pdtc-1.0-${t,,}-m/files_train
ls data/$t/pml/*_{[0-9],[0-9][0-9]}6.*m.gz | sed 's/\.[am]\.gz$//' > pdtc-1.0-${t,,}-m/files_dev
ls data/$t/pml/*_{[0-9],[0-9][0-9]}7.*m.gz | sed 's/\.[am]\.gz$//' > pdtc-1.0-${t,,}-m/files_test
done
for d in pdtc-1.0-*/; do
cat $d/files_train $d/files_dev $d/files_test >$d/files_all
done