-
Notifications
You must be signed in to change notification settings - Fork 29
/
val.py
105 lines (90 loc) · 3.61 KB
/
val.py
1
2
3
4
5
6
7
8
9
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
42
43
44
45
46
47
48
49
50
51
52
53
54
55
56
57
58
59
60
61
62
63
64
65
66
67
68
69
70
71
72
73
74
75
76
77
78
79
80
81
82
83
84
85
86
87
88
89
90
91
92
93
94
95
96
97
98
99
100
101
102
103
104
105
import bilstm_crf_model
import process_data
import numpy as np
import pandas as pd
import argparse
from sklearn.metrics import classification_report
from utils import *
from tqdm import tqdm
def get_test_data(txt_data, tag_data):
with open(txt_data,'rb') as f:
predict_text = f.read().decode('utf-8')
predict_list = [char for char in predict_text]
char_num = len(predict_list)
tag = pd.read_csv(tag_data, header=None, sep='\t')
tag_list = ['O' for _ in range( char_num )]
for i in range(tag.shape[0]):
tag_item = tag.iloc[i][1].split(' ')
cls, start, end = tag_item[0], int( tag_item[1] ), int( tag_item[-1] )
tag_list[start] = 'B-'+cls
for j in range(start+1, end):
tag_list[j] = 'I-'+cls
return predict_text, tag_list
def show(predict_text, result_tags, cls_name='Disease'):
result = ''
for s, t in zip(predict_text, result_tags):
if t in ('B-'+cls_name, 'I-'+cls_name):
result += ' ' + s if (t == 'B-'+cls_name) else s
print([cls_name+':' + result])
def local_test(txt_data, tag_data, model, word2idx, chunk_tags):
predict_text, tag_list = get_test_data(txt_data, tag_data)
str_, length = process_data.process_data(predict_text, word2idx, len(tag_list))
raw = model.predict(str_)[0][-length:]
result = [np.argmax(row) for row in raw]
result_tags = [chunk_tags[i] for i in result]
print(classification_report(tag_list, result_tags))
#show(predict_text, result_tags, "Drug")
def savefile(tag_data,result_tags, predict_text):
tags = []
for tag in result_tags:
if tag != 'O':
tag_ = tag.split('-')[1]
else:
tag_ = tag
tags.append(tag_)
# write here
prev = tags[0]
start = 0
num = 0
for i in range(1, len(tags)):
cur = tags[i]
if cur != prev:
end = i
if prev != 'O':
num += 1
content = predict_text[start:end]
content = content.replace('\n',' ')
tag_data.write('T'+str(num)+'\t'+prev+' '+str(start)+' '+str(end)+'\t'+content+'\n')
start = i
prev = cur
tag_data.close()
def test(test_dir, submit_dir, model, word2idx, chunk_tags):
for filename in tqdm( os.listdir(test_dir) ):
fileidx = filename.split('.')[0]
txt_data = test_dir + filename
tag_data = open(submit_dir+fileidx+'.ann','w')
with open(txt_data,'rb') as f:
predict_text = f.read().decode('utf-8')
str_, length = process_data.process_data(predict_text, word2idx, len(predict_text))
raw = model.predict(str_)[0][-length:]
result = [np.argmax(row) for row in raw]
result_tags = [chunk_tags[i] for i in result]
savefile(tag_data, result_tags, predict_text)
if __name__ == '__main__':
parser = argparse.ArgumentParser(description='VAL')
parser.add_argument('--num', type=int)
parser.add_argument('--embed', type=int)
parser.add_argument('--units', type=int)
parser.add_argument('--gpu', type=int)
args = parser.parse_args()
gpu_config(args.gpu)
# load model
model_dir = 'expr/'+str(args.num)+'/model.h5'
model, (word2idx, chunk_tags) = bilstm_crf_model.create_model(args.embed, args.units, train=False)
model.load_weights(model_dir)
test_dir = 'data/raw/test/'
submit_dir = 'data/raw/submit/'
#test(test_dir, submit_dir, model,word2idx, chunk_tags)
txt_data = 'data/raw/local_test/152_6.txt'
tag_data = 'data/raw/local_test/152_6.ann'
local_test(txt_data, tag_data,model, word2idx, chunk_tags)