-
Notifications
You must be signed in to change notification settings - Fork 0
/
Bayes.py
96 lines (65 loc) · 2.48 KB
/
Bayes.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
from emailClass import *
from numpy import *
mk = 1
bk = 1
def multinomial(classific):
# number of times word appears in document/ documents in class
for doc in classific.training_data:
# for each word in the doc
for key,value in doc.dictionary.iteritems():
if(key not in classific.m_likelihood):
counter = 0
# for each doc in the training data
for doc in classific.training_data:
if(key in doc.dictionary):
counter +=doc.dictionary[key]
classific.m_likelihood[key] = float(counter + mk)/(2*mk+classific.total_words)
def bernouilli(classific):
# number of documents word appears/ documents in class
# for every doc in the training data
for doc in classific.training_data:
# for each word in the doc
for key,value in doc.dictionary.iteritems():
if(key not in classific.b_likelihood):
counter = 0
# for each doc in the training data
for doc in classific.training_data:
if(key in doc.dictionary):
counter +=1
classific.b_likelihood[key] = float(counter+bk)/(2*bk +len(classific.training_data))
def multinomial_map_estimate(testdoc, emailclass, otherclass):
map_estimate = 0.0
for key in testdoc.dictionary:
if(key in emailclass.m_likelihood):
map_estimate = float(map_estimate) + math.log(emailclass.m_likelihood[key])
elif(key in otherclass.m_likelihood):
guess = float(mk)/emailclass.total_words
map_estimate += math.log(guess)
return map_estimate
def multinomial_classifier(testdoc, spamclass, regclass):
list = []
map_spam = multinomial_map_estimate(testdoc, spamclass, regclass)
map_reg = multinomial_map_estimate(testdoc, regclass, spamclass)
list.append(map_reg)
list.append(map_spam)
testdoc.set_label(list.index(max(list)))
return max(map_reg, map_spam), list.index(max(list))
def bernouilli_map_estimate(testdoc, emailclass, otherclass):
map_estimate = 0.0
for key in testdoc.dictionary:
if(key in emailclass.b_likelihood):
map_estimate += float(math.log(emailclass.b_likelihood[key]))
elif(key in otherclass.b_likelihood):
guess = float(bk)/emailclass.total_words
map_estimate += math.log(guess)
return map_estimate
def bernouilli_classifier(testdoc, spamclass, regclass):
list = []
map_spam = bernouilli_map_estimate(testdoc, spamclass, regclass)
map_reg = bernouilli_map_estimate(testdoc, regclass, spamclass)
list.append(map_reg)
list.append(map_spam)
testdoc.set_label(list.index(max(list)))
return max(list), list.index(max(list))
if __name__ == '__main__':
main()