You can not select more than 25 topics Topics must start with a letter or number, can include dashes ('-') and can be up to 35 characters long.
cjvt-srl-tagging/tools/check_all_files_existence.py

20 lines
833 B

import os
# INPATH = Path(config["tools"]["giga_srl"])
# infiles = list(enumerate([x for x in iter(sorted(INPATH.iterdir())) if x.is_file()]))
SOURCEPATH = '/media/luka/Portable Disk/Datasets/gigafida_jos/gf2_files'
from shutil import copyfile
INPATH = '/media/luka/Portable Disk/Datasets/gigafida_jos/2_srl'
OUTPATH = '/home/luka/Development/srl/cjvt-srl-tagging/data/giga_out/1_tsv'
for i in range(100000):
# print(os.path.join(INPATH, 'giga.%07d.tsv' % i))
# if not os.path.exists(os.path.join(INPATH, 'giga.%07d.tsv' % i)):
# print('giga.%07d.tsv' % i)
if not os.path.exists(os.path.join(INPATH, 'giga%07d.srl.tsv' % i)):
copyfile(os.path.join(SOURCEPATH, 'giga.%07d.tsv' % i), os.path.join(OUTPATH, 'giga%07d.tsv' % i))
print('giga%07d.srl.tsv' % i)
if i % 1000 == 0:
print(i)