-
Notifications
You must be signed in to change notification settings - Fork 0
/
Copy pathfasttext_context_class.py
187 lines (174 loc) · 8.47 KB
/
fasttext_context_class.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
126
127
128
129
130
131
132
133
134
135
136
137
138
139
140
141
142
143
144
145
146
147
148
149
150
151
152
153
154
155
156
157
158
159
160
161
162
163
164
165
166
167
168
169
170
171
172
173
174
175
176
177
178
179
180
181
182
183
from context_queue import robot_context_queue
from jieba_cut import *
import fastText as fasttext
import re
from keyword_nv import robot_keyword_nv
class Fasttext_Context(object):
"""
功能：智能客服语义分析引擎-fasttext上下文分类
输入：
输出：
方法：构建存储上下文的队列
"""
def __init__(self,
threshold=0.8 # 分类阈值
):
### STEP1：定义初始化数据 ###
self.threshold = threshold
self.fasttext_model=fasttext.load_model('./fasttext_model/sn_cate/classf_knowledge.model')
def outstr_str(self,text):
outstr = ''
for word in jieba_cut(text):
if word != '\t':
outstr += word
outstr += " "
return outstr
def fasttext_single(self,text):
outstr = self.outstr_str(text)
cate = self.fasttext_model.predict(outstr, k=1)
cate=re.sub('__label__', '', cate[0][0])
return cate
def fasttext_context(self,Text, userId, requesId):
robot_context_queue.context_queue(Text, userId, requesId)
print("robot_context_queue.UserText", robot_context_queue.UserText)
len_userId_UserText = len(robot_context_queue.UserText[userId])
if len_userId_UserText == 1:
text_0 = robot_context_queue.UserText[userId][len_userId_UserText - 1][0] # 当前输入值
outstr_0 = self.outstr_str(text_0)
cate_0 = self.fasttext_model.predict(outstr_0, k=2)
# print("outstr_0", outstr_0)
# print("cate_0", cate_0)
cate = re.sub('__label__', '', cate_0[0][0])
return cate
elif len_userId_UserText == 2:
text_0 = robot_context_queue.UserText[userId][len_userId_UserText - 1][0] # 当前输入值
text_1 = robot_context_queue.UserText[userId][len_userId_UserText - 2][0] # 前一条数据
outstr_0 = self.outstr_str(text_0)
cate_0 = self.fasttext_model.predict(outstr_0, k=2)
# print("outstr_0", outstr_0)
# print("cate_0", cate_0)
if cate_0[1][0] >= self.threshold:
cate = re.sub('__label__', '', cate_0[0][0])
return cate
else:
text_01 = text_1 + text_0
outstr_01 = self.outstr_str(text_01)
cate_01 = self.fasttext_model.predict(outstr_01, k=2)
# print("outstr_01", outstr_01)
# print("cate_01", cate_01)
cate = re.sub('__label__', '', cate_01[0][0])
return cate
elif len_userId_UserText == 3:
text_0 = robot_context_queue.UserText[userId][len_userId_UserText - 1][0] # 当前输入值
text_1 = robot_context_queue.UserText[userId][len_userId_UserText - 2][0] # 前一条数据
text_2 = robot_context_queue.UserText[userId][len_userId_UserText - 3][0] # 前前一条数据
outstr_0 = self.outstr_str(text_0)
cate_0 = self.fasttext_model.predict(outstr_0, k=2)
# print("outstr_0", outstr_0)
# print("cate_0", cate_0)
if cate_0[1][0] >= self.threshold:
cate = re.sub('__label__', '', cate_0[0][0])
return cate
else:
text_01 = text_1 + text_0
outstr_01 = self.outstr_str(text_01)
cate_01 = self.fasttext_model.predict(outstr_01, k=2)
# print("outstr_01", outstr_01)
# print("cate_01", cate_01)
if cate_01[1][0] >= self.threshold:
cate = re.sub('__label__', '', cate_01[0][0])
return cate
else:
text_012 = text_2 + text_1 + text_0
outstr_012 = self.outstr_str(text_012)
cate_012 = self.fasttext_model.predict(outstr_012, k=2)
# print("outstr_012", outstr_012)
# print("cate_012", cate_012)
cate = re.sub('__label__', '', cate_012[0][0])
return cate
else:
cate = self.fasttext_single(Text)
return cate
def fasttext_keyword_single(self,text):
outstr=robot_keyword_nv.keyword_nv(text)
# outstr = self.outstr_str(keyword_str)
cate = self.fasttext_model.predict(outstr, k=1)
cate=re.sub('__label__', '', cate[0][0])
return cate
def fasttext_keyword_context(self,Text, userId, requesId):
robot_context_queue.context_queue(Text, userId, requesId)
# print("robot_context_queue.UserText", robot_context_queue.UserText)
len_userId_UserText = len(robot_context_queue.UserText[userId])
if len_userId_UserText == 1:
text_0 = robot_context_queue.UserText[userId][len_userId_UserText - 1][0] # 当前输入值
outstr_0 = robot_keyword_nv.keyword_nv(text_0)
# outstr_0 = self.outstr_str(text_0)
cate_0 = self.fasttext_model.predict(outstr_0, k=2)
# print("outstr_0", outstr_0)
# print("cate_0", cate_0)
cate = re.sub('__label__', '', cate_0[0][0])
return cate
elif len_userId_UserText == 2:
text_0 = robot_context_queue.UserText[userId][len_userId_UserText - 1][0] # 当前输入值
text_1 = robot_context_queue.UserText[userId][len_userId_UserText - 2][0] # 前一条数据
outstr_0 = robot_keyword_nv.keyword_nv(text_0)
# outstr_0 = self.outstr_str(text_0)
cate_0 = self.fasttext_model.predict(outstr_0, k=2)
# print("outstr_0", outstr_0)
# print("cate_0", cate_0)
if cate_0[1][0] >= self.threshold:
cate = re.sub('__label__', '', cate_0[0][0])
return cate
else:
text_01 = text_1 + text_0
outstr_01 = robot_keyword_nv.keyword_nv(text_01)
# outstr_01 = self.outstr_str(text_01)
cate_01 = self.fasttext_model.predict(outstr_01, k=2)
# print("outstr_01", outstr_01)
# print("cate_01", cate_01)
cate = re.sub('__label__', '', cate_01[0][0])
return cate
elif len_userId_UserText == 3:
text_0 = robot_context_queue.UserText[userId][len_userId_UserText - 1][0] # 当前输入值
text_1 = robot_context_queue.UserText[userId][len_userId_UserText - 2][0] # 前一条数据
text_2 = robot_context_queue.UserText[userId][len_userId_UserText - 3][0] # 前前一条数据
outstr_0 = robot_keyword_nv.keyword_nv(text_0)
# outstr_0 = self.outstr_str(text_0)
cate_0 = self.fasttext_model.predict(outstr_0, k=2)
# print("outstr_0", outstr_0)
# print("cate_0", cate_0)
if cate_0[1][0] >= self.threshold:
cate = re.sub('__label__', '', cate_0[0][0])
return cate
else:
text_01 = text_1 + text_0
outstr_01 = robot_keyword_nv.keyword_nv(text_01)
# outstr_01 = self.outstr_str(text_01)
cate_01 = self.fasttext_model.predict(outstr_01, k=2)
# print("outstr_01", outstr_01)
# print("cate_01", cate_01)
if cate_01[1][0] >= self.threshold:
cate = re.sub('__label__', '', cate_01[0][0])
return cate
else:
text_012 = text_2 + text_1 + text_0
outstr_012 = robot_keyword_nv.keyword_nv(text_012)
# outstr_012 = self.outstr_str(text_012)
cate_012 = self.fasttext_model.predict(outstr_012, k=2)
# print("outstr_012", outstr_012)
# print("cate_012", cate_012)
cate = re.sub('__label__', '', cate_012[0][0])
return cate
else:
cate = self.fasttext_keyword_single(Text)
return cate
if __name__=='__main__':
robot_fasttext_context=Fasttext_Context()
while True:
text=input("请输入测试问题：")
userid=input("请输入用户编号：")
requid=input("请输入session代码：")
# cate = robot_fasttext_context.fasttext_context(text, userid, requid)
# cate = robot_fasttext_context.fasttext_keyword_single(text)
cate = robot_fasttext_context.fasttext_keyword_context(text, userid, requid)
print(cate)