-
Notifications
You must be signed in to change notification settings - Fork 1
/
mapping_tools.py
126 lines (100 loc) · 4.29 KB
/
mapping_tools.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
106
107
108
109
110
111
112
113
114
115
116
117
118
119
120
121
122
123
124
125
from collections import namedtuple
import sys
import xml
import utils
import re
Word = namedtuple('WordWithConfidence', ['word', 'confidence'])
class QueryContentHandler(utils.AutoContentHandler):
'''
Example XML
<parameters>
<query>
<number>75739</number>
<text> #weight(1.0 variant.(class) 1.0 has.(class) 1.0 string.(class)) </text>
</query>
</parameters>
'''
queryWordPattern = r"\d+\.\d+\s+(.+?)\.\((.+?)\)"
def __init__(self, wordMapper, confidenceThreshold):
self.bugRepositoryName = None
self.currentBugReportId = None
self.currentBugReportWords = []
self.currentFiles = []
self.bugReportInformation = []
self.lastTag = None
self.wordMapper = wordMapper
self.confidenceThreshold = confidenceThreshold
def start_parameters(self, name, attrs):
sys.stdout.write("<parameters>\n")
def start_query(self, name, attrs):
sys.stdout.write("\t<query>\n")
def start_number(self, name, attrs):
self.lastTag = name
sys.stdout.write("\t\t<number>")
def start_text(self, name, attrs):
self.lastTag = name
sys.stdout.write("\t\t<text> #weight(")
def end_parameters(self, name):
sys.stdout.write("</parameters>\n")
def end_query(self, name):
sys.stdout.write("\t</query>\n")
def end_number(self, name):
sys.stdout.write("</number>\n")
def end_text(self, name):
sys.stdout.write(") </text>\n")
def characters(self, content):
if self.lastTag == "number":
sys.stdout.write(content)
elif self.lastTag == "text":
sys.stdout.write(self.mapWords(content))
def mapWords(self, wordContent):
return \
" ".join(
[
"1.0 %s.(%s)" % (word, match.group(2))
for match in re.finditer(QueryContentHandler.queryWordPattern, wordContent)
for word in self.wordMapper.mapWord(match.group(1), match.group(2), self.confidenceThreshold)
]
)
class WordMapper:
def __init__(self, wordMappingsInContext):
self.wordMappingsInContext = wordMappingsInContext
def mapWord(self, word, context, confidenceThreshold):
return [word] if word not in self.wordMappingsInContext[context] else \
map(
lambda x: x.word,
filter(lambda x: x.confidence > confidenceThreshold, self.wordMappingsInContext[context][word])
)
def createWordMappingFromModel(contextToWordModelMap, sourceCorpus, targetCorpus):
wordMapping = { context : {} for context in contextToWordModelMap }
for context in contextToWordModelMap:
for sourceWord in sourceCorpus.vocabulary:
sourceWordVector = sourceCorpus.convertToWordVector([sourceWord])
targetWordVector = ae.predict(np.asarray([sourceWordVector], dtype=np.float32))[0]
sortedIndices = np.argsort(targetWordVector)[::-1] # high confidence to low confidence
targetWords = map(lambda x: Word(x), zip(np.asarray(targetCorpus.vocabulary)[sortedIndices], targetWordVector[sortedIndices]))
wordMapping[context][sourceWord] = targetWords
return wordMapping
if __name__ == "__main__":
# Example code
# John provides ...
wordMapping_normalContext = {
"cat": [Word("mouse", 0.9), Word("hairball", 0.4), Word("bird", 0.7)],
"dog": [Word("bone", .99), Word("hammer", 0.01), Word("meat", 0.8), Word("bath", 0.4)],
"cow": [Word("spot", .79), Word("milk", 1.0)]
}
wordMapping_weirdContext = {
"cat": [Word("planet", 0.3), Word("coin", 0.43), Word("arrow", 0.92)],
"dog": [Word("fork", .95), Word("foil", 0.51)],
"cow": [Word("hair", .43), Word("dirt", 0.02), Word("bottle", 0.74), Word("light", 0.24)]
}
# Vince provides ...
wordMapper = WordMapper(
{ "normal": wordMapping_normalContext, "weird": wordMapping_weirdContext }
)
print wordMapper.mapWord("cat", "normal", 0.75)
print wordMapper.mapWord("dog", "normal", 0.75)
print wordMapper.mapWord("cow", "normal", 0.75)
print wordMapper.mapWord("cat", "weird", 0.48)
print wordMapper.mapWord("dog", "weird", 0.48)
print wordMapper.mapWord("cow", "weird", 0.48)