Created
July 3, 2019 07:23
-
-
Save Taekyoon/e7cf95b7895b065332c92f74744dae46 to your computer and use it in GitHub Desktop.
sequence tagging utils
This file contains hidden or bidirectional Unicode text that may be interpreted or compiled differently than what appears below. To review, open the file in an editor that reveals hidden Unicode characters.
Learn more about bidirectional Unicode characters
def base_to_char_level_bio_tags(sent, tag, empty_tag='O'): | |
char_level_tags = list() | |
sent_list, tag_list = sent.split(), tag.split() | |
if len(sent_list) != len(tag_list): | |
raise ValueError() | |
for i, (s, t) in enumerate(zip(sent_list, tag_list)): | |
char_level_tags += [t for _ in range(len(s))] | |
if i >= len(sent_list) - 1: | |
break | |
if tag[i] == tag[i+1]: | |
char_level_tags += [t] | |
else: | |
char_level_tags += [empty_tag] | |
return ' '.join(char_level_tags) | |
def transit_bio_tags_level_by_sent(sent, label): | |
transited_label = list() | |
label_pos = 0 | |
for t in sent.split(): | |
t_l = len(t) | |
next_label_pos = label_pos + t_l | |
label_seq = label[label_pos:next_label_pos] | |
target_label = 'O' | |
for l in label_seq: | |
if 'B' in l: | |
target_label = l | |
break | |
if 'I' in l: | |
target_label = l | |
transited_label.append(target_label) | |
label_pos = next_label_pos | |
return ' '.join(transited_label) |
Sign up for free
to join this conversation on GitHub.
Already have an account?
Sign in to comment