-
Notifications
You must be signed in to change notification settings - Fork 0
/
loc_features.py
46 lines (40 loc) · 1.12 KB
/
loc_features.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
from sklearn.feature_extraction.text import CountVectorizer
def preprocess(name):
strip = ['(',')','-']
name = '$' + (name.replace(' ', '$')).lower() + '$'
for s in strip:
name = name.replace(s, '')
return name
def build_features(types, names):
vector = []
for name in names:
name = preprocess(name)
if types == 3:
grams = find_trigrams(name)
else:
grams = find_bigrams(name)
for g in grams:
if g not in vector:
vector.append(g)
return vector
def get_features(features, name):
grams = find_trigrams(preprocess(name))
vector = [0] * len(features)
for gram in grams:
try:
idx = features.index(gram)
except:
print('not found')
else:
vector[idx] = 1
return vector
def find_trigrams(name):
trigrams = []
for i in range(len(name)-2):
trigrams.append(name[i] + name[i+1] + name[i+2])
return trigrams
def find_bigrams(name):
bigrams = []
for i in range(len(name) - 1):
bigrams.append(name[i] + name[i + 1])
return bigrams