from baseline.utils import convert_iob_to_bio
import os
import argparse
parser = argparse.ArgumentParser(
description='Translate input sequence to output sequence')
parser.add_argument('--io_dir', help='Input/Output dir', default='../data')
parser.add_argument('--train_file',
help='Training file relative name',
default='eng.train')
parser.add_argument('--valid_file',
help='Validation file relative name',
default='eng.testa')
parser.add_argument('--test_file',
help='Test file relative name',
default='eng.testb')
parser.add_argument('--suffix', help='Suffix to append', default='.bio')
args = parser.parse_args()
convert_iob_to_bio(os.path.join(args.io_dir, args.train_file),
os.path.join(args.io_dir, args.train_file + args.suffix))
convert_iob_to_bio(os.path.join(args.io_dir, args.valid_file),
os.path.join(args.io_dir, args.valid_file + args.suffix))
convert_iob_to_bio(os.path.join(args.io_dir, args.test_file),
os.path.join(args.io_dir, args.test_file + args.suffix))
def test():
spans = generate_spans()
iob = generate_iob(spans)
gold_bio = generate_bio(spans)
bio = convert_iob_to_bio(iob)
assert bio == gold_bio
def test_iob_bio_i_first():
in_ = ['I-X', 'O']
gold = ['B-X', 'O']
res = convert_iob_to_bio(in_)
assert res == gold
def test_iob_bio_i_after_i_same():
in_ = ['O', 'I-X', 'I-X', 'O']
gold = ['O', 'B-X', 'I-X', 'O']
res = convert_iob_to_bio(in_)
assert res == gold
def test_iob_bio_i_after_b_diff():
in_ = ['O', 'B-X', 'I-Y', 'O']
gold = ['O', 'B-X', 'B-Y', 'O']
res = convert_iob_to_bio(in_)
assert res == gold
def test():
spans = generate_spans()
gold_bio = generate_bio(spans)
res = convert_iob_to_bio(convert_bio_to_iob(gold_bio))
assert res == gold_bio
