Spaces:
Sleeping
Sleeping
naveenvenkatesh
commited on
Commit
•
a1634d3
1
Parent(s):
dc08b8c
Update app.py
Browse files
app.py
CHANGED
@@ -12,12 +12,13 @@ from transformers import pipeline
|
|
12 |
import matplotlib.pyplot as plt
|
13 |
import plotly.express as px
|
14 |
|
|
|
|
|
15 |
class SentimentAnalyzer:
|
16 |
def __init__(self):
|
17 |
-
|
18 |
self.model="facebook/bart-large-mnli"
|
19 |
|
20 |
-
self.client = OpenAI()
|
21 |
|
22 |
def analyze_sentiment(self, text):
|
23 |
pipe = pipeline("zero-shot-classification", model=self.model)
|
@@ -28,17 +29,24 @@ class SentimentAnalyzer:
|
|
28 |
return sentiment_scores_str
|
29 |
|
30 |
def emotion_analysis(self,text):
|
31 |
-
|
32 |
-
|
33 |
-
|
34 |
-
|
35 |
-
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
36 |
temperature=1,
|
37 |
max_tokens=60
|
38 |
)
|
39 |
|
40 |
-
message = response.choices[0].
|
41 |
-
print(message)
|
42 |
return message
|
43 |
|
44 |
def analyze_sentiment_for_graph(self, text):
|
@@ -55,8 +63,8 @@ Please analyze the text and provide the output in the following format: emotion:
|
|
55 |
def emotion_analysis_for_graph(self,text):
|
56 |
|
57 |
list_of_emotion=text.split(":")
|
58 |
-
label=list_of_emotion[
|
59 |
-
score=list_of_emotion[
|
60 |
score_dict={
|
61 |
label:float(score)
|
62 |
}
|
@@ -67,21 +75,26 @@ Please analyze the text and provide the output in the following format: emotion:
|
|
67 |
class Summarizer:
|
68 |
|
69 |
def __init__(self):
|
70 |
-
self.client = OpenAI()
|
|
|
71 |
|
72 |
def generate_summary(self, text):
|
73 |
-
|
74 |
-
|
|
|
|
|
75 |
write within 30 words.
|
76 |
-
```{text}```
|
77 |
-
|
78 |
-
|
79 |
-
|
80 |
-
|
81 |
-
|
82 |
-
temperature=
|
|
|
83 |
)
|
84 |
-
|
|
|
85 |
return message
|
86 |
|
87 |
history_state = gr.State()
|
@@ -91,20 +104,33 @@ sentiment = SentimentAnalyzer()
|
|
91 |
class LangChain_Document_QA:
|
92 |
|
93 |
def __init__(self):
|
94 |
-
self.client = OpenAI()
|
|
|
95 |
|
96 |
def _add_text(self,history, text):
|
|
|
97 |
history = history + [(text, None)]
|
98 |
history_state.value = history
|
99 |
-
return history,gr.update(value="", interactive=False)
|
100 |
|
101 |
-
|
102 |
-
|
103 |
-
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
104 |
history_state.value = history
|
|
|
|
|
|
|
105 |
return history
|
106 |
|
|
|
107 |
def _chat_history(self):
|
|
|
108 |
history = history_state.value
|
109 |
formatted_history = " "
|
110 |
for entry in history:
|
@@ -142,31 +168,39 @@ class LangChain_Document_QA:
|
|
142 |
return client,agent
|
143 |
|
144 |
|
145 |
-
def _suggested_answer(self,text):
|
146 |
-
try:
|
147 |
-
|
148 |
-
|
149 |
-
|
150 |
-
|
|
|
151 |
file_path = "vodafone_customer_details.json"
|
152 |
with open(file_path) as file:
|
153 |
-
|
154 |
-
|
155 |
-
|
156 |
-
|
157 |
-
|
158 |
-
|
159 |
-
|
|
|
|
|
160 |
temperature=0,
|
161 |
-
max_tokens=
|
162 |
)
|
163 |
-
|
164 |
-
message = response.choices[0].
|
|
|
|
|
165 |
if ":" in message:
|
166 |
message = re.sub(r'^.*:', '', message)
|
167 |
-
|
168 |
-
|
169 |
-
|
|
|
|
|
|
|
170 |
|
171 |
|
172 |
def _text_box(self,customer_emotion,agent_emotion,agent_sentiment_score,customer_sentiment_score):
|
@@ -205,7 +239,7 @@ class LangChain_Document_QA:
|
|
205 |
|
206 |
|
207 |
def clear_func(self):
|
208 |
-
history_state.
|
209 |
|
210 |
|
211 |
def gradio_interface(self):
|
@@ -217,7 +251,7 @@ class LangChain_Document_QA:
|
|
217 |
gr.HTML("""<center><h1 style="color:#f26d25">Generative AI CRM ChatBot</h1></center>""")
|
218 |
with gr.Row():
|
219 |
gr.HTML("<br>")
|
220 |
-
chatbot = gr.Chatbot(
|
221 |
with gr.Row():
|
222 |
with gr.Column(scale=0.50):
|
223 |
txt = gr.Textbox(
|
@@ -280,13 +314,12 @@ class LangChain_Document_QA:
|
|
280 |
with gr.Column(scale=0.70, min_width=0):
|
281 |
plot_4 =gr.Plot(label="Agent_Emotion", size=(500, 600))
|
282 |
|
283 |
-
|
284 |
txt_msg = txt.submit(self._add_text, [chatbot, txt], [chatbot, txt])
|
285 |
-
txt_msg.then(lambda: gr.update(interactive=True), None, [txt])
|
286 |
-
txt.submit(self._suggested_answer,txt,txt3)
|
287 |
-
button.click(self._agent_text, [chatbot
|
288 |
-
txt2.submit(self._agent_text, [chatbot
|
289 |
-
self._agent_text, [chatbot
|
290 |
)
|
291 |
end_btn.click(self._display_history, [], txt4)
|
292 |
emptyBtn.click(self.clear_func,[],[])
|
|
|
12 |
import matplotlib.pyplot as plt
|
13 |
import plotly.express as px
|
14 |
|
15 |
+
# client = OpenAI()
|
16 |
+
|
17 |
class SentimentAnalyzer:
|
18 |
def __init__(self):
|
|
|
19 |
self.model="facebook/bart-large-mnli"
|
20 |
|
21 |
+
# self.client = OpenAI()
|
22 |
|
23 |
def analyze_sentiment(self, text):
|
24 |
pipe = pipeline("zero-shot-classification", model=self.model)
|
|
|
29 |
return sentiment_scores_str
|
30 |
|
31 |
def emotion_analysis(self,text):
|
32 |
+
client = OpenAI()
|
33 |
+
print(text,"lkdjrglk")
|
34 |
+
conversation = [
|
35 |
+
{"role": "system", "content": """You are a Emotion Analyser.Your task is to analyze and predict the emotion using scores. Emotions are categorized into the following list: Sadness, Happiness, Joy, Fear, Disgust, and Anger. You need to provide the emotion with the highest score. The scores should be in the range of 0.0 to 1.0, where 1.0 represents the highest intensity of the emotion.
|
36 |
+
Please analyze the text and provide the output in the following format: emotion: score [with one result having the highest score]."""},
|
37 |
+
{"role": "user", "content": f"""
|
38 |
+
input text{text}
|
39 |
+
"""}
|
40 |
+
]
|
41 |
+
response = client.chat.completions.create(
|
42 |
+
model="gpt-3.5-turbo",
|
43 |
+
messages=conversation,
|
44 |
temperature=1,
|
45 |
max_tokens=60
|
46 |
)
|
47 |
|
48 |
+
message = response.choices[0].message.content
|
49 |
+
print("sen",message)
|
50 |
return message
|
51 |
|
52 |
def analyze_sentiment_for_graph(self, text):
|
|
|
63 |
def emotion_analysis_for_graph(self,text):
|
64 |
|
65 |
list_of_emotion=text.split(":")
|
66 |
+
label=list_of_emotion[1]
|
67 |
+
score=list_of_emotion[2]
|
68 |
score_dict={
|
69 |
label:float(score)
|
70 |
}
|
|
|
75 |
class Summarizer:
|
76 |
|
77 |
def __init__(self):
|
78 |
+
# self.client = OpenAI()
|
79 |
+
pass
|
80 |
|
81 |
def generate_summary(self, text):
|
82 |
+
client = OpenAI()
|
83 |
+
conversation = [
|
84 |
+
{"role": "system", "content": "You are a Summarizer"},
|
85 |
+
{"role": "user", "content": f"""summarize the following conversation delimited by triple backticks.
|
86 |
write within 30 words.
|
87 |
+
```{text}```
|
88 |
+
"""}
|
89 |
+
]
|
90 |
+
response = client.chat.completions.create(
|
91 |
+
model="gpt-3.5-turbo",
|
92 |
+
messages=conversation,
|
93 |
+
temperature=1,
|
94 |
+
max_tokens=100
|
95 |
)
|
96 |
+
|
97 |
+
message = response.choices[0].message.content
|
98 |
return message
|
99 |
|
100 |
history_state = gr.State()
|
|
|
104 |
class LangChain_Document_QA:
|
105 |
|
106 |
def __init__(self):
|
107 |
+
# self.client = OpenAI()
|
108 |
+
pass
|
109 |
|
110 |
def _add_text(self,history, text):
|
111 |
+
|
112 |
history = history + [(text, None)]
|
113 |
history_state.value = history
|
|
|
114 |
|
115 |
+
return history, gr.Textbox(value="", interactive=False)
|
116 |
+
|
117 |
+
def _agent_text(self,text,history):
|
118 |
+
# respond = []
|
119 |
+
# print("history",history)
|
120 |
+
# print(type(history))
|
121 |
+
# text = [text for _, text in history][-1]
|
122 |
+
# print("agent",history_state.value)
|
123 |
+
|
124 |
+
history[-1][1] = text
|
125 |
history_state.value = history
|
126 |
+
# message = [tup[-1] for tup in history]
|
127 |
+
# respond.append(("", message))
|
128 |
+
# print(respond)
|
129 |
return history
|
130 |
|
131 |
+
|
132 |
def _chat_history(self):
|
133 |
+
print("chat history",history_state.value)
|
134 |
history = history_state.value
|
135 |
formatted_history = " "
|
136 |
for entry in history:
|
|
|
168 |
return client,agent
|
169 |
|
170 |
|
171 |
+
def _suggested_answer(self,history,text):
|
172 |
+
# try:
|
173 |
+
client = OpenAI()
|
174 |
+
self._add_text(history,text)
|
175 |
+
text_history = self._chat_history()
|
176 |
+
# start_sequence = "\nCustomer:"
|
177 |
+
# restart_sequence = "\nVodafone Customer Care:"
|
178 |
file_path = "vodafone_customer_details.json"
|
179 |
with open(file_path) as file:
|
180 |
+
context = json.load(file)
|
181 |
+
conversation = [
|
182 |
+
{"role": "system", "content": f"You Are Vodafone Sim AI Chatbot. Use the following pieces of context{context} to answer the question at the end. If you don't know the answer, just say that you don't know, don't try to make up an answer. Use three sentences maximum. Keep the answer as concise as possible. if user say hi, hello you say welcome greetings like hi, hello. if user say thankyou, thanks tone you say thanking Greetings like You're welcome!.conversation history{context}"},
|
183 |
+
{"role": "user", "content": f""" this is the user question:{text}..
|
184 |
+
"""}
|
185 |
+
]
|
186 |
+
response = client.chat.completions.create(
|
187 |
+
model="gpt-3.5-turbo",
|
188 |
+
messages=conversation,
|
189 |
temperature=0,
|
190 |
+
max_tokens=100
|
191 |
)
|
192 |
+
|
193 |
+
message = response.choices[0].message.content
|
194 |
+
print("message",message)
|
195 |
+
|
196 |
if ":" in message:
|
197 |
message = re.sub(r'^.*:', '', message)
|
198 |
+
history.append((text, message))
|
199 |
+
else:
|
200 |
+
history.append((text, message))
|
201 |
+
return "",message
|
202 |
+
# except:
|
203 |
+
# return "Hi, How can I help you?"
|
204 |
|
205 |
|
206 |
def _text_box(self,customer_emotion,agent_emotion,agent_sentiment_score,customer_sentiment_score):
|
|
|
239 |
|
240 |
|
241 |
def clear_func(self):
|
242 |
+
history_state = gr.State([])
|
243 |
|
244 |
|
245 |
def gradio_interface(self):
|
|
|
251 |
gr.HTML("""<center><h1 style="color:#f26d25">Generative AI CRM ChatBot</h1></center>""")
|
252 |
with gr.Row():
|
253 |
gr.HTML("<br>")
|
254 |
+
chatbot = gr.Chatbot().style(height=300)
|
255 |
with gr.Row():
|
256 |
with gr.Column(scale=0.50):
|
257 |
txt = gr.Textbox(
|
|
|
314 |
with gr.Column(scale=0.70, min_width=0):
|
315 |
plot_4 =gr.Plot(label="Agent_Emotion", size=(500, 600))
|
316 |
|
|
|
317 |
txt_msg = txt.submit(self._add_text, [chatbot, txt], [chatbot, txt])
|
318 |
+
# txt_msg.then(lambda: gr.update(interactive=True), None, [txt])
|
319 |
+
txt.submit(self._suggested_answer, [chatbot, txt], [txt,txt3])
|
320 |
+
button.click(self._agent_text, [txt3,chatbot], chatbot)
|
321 |
+
txt2.submit(self._agent_text, [txt2,chatbot ], chatbot).then(
|
322 |
+
self._agent_text, [txt2,chatbot], chatbot
|
323 |
)
|
324 |
end_btn.click(self._display_history, [], txt4)
|
325 |
emptyBtn.click(self.clear_func,[],[])
|