VectorV17 commited on
Commit
e5f0bd4
·
verified ·
1 Parent(s): 3e15df8

Update app.py

Browse files
Files changed (1) hide show
  1. app.py +12 -31
app.py CHANGED
@@ -16,7 +16,6 @@ load_dotenv()
16
  # -------------------------------------------------------------------------
17
  # 1. Mock Data Setup (Simplilearn MOOC FAQs)
18
  # -------------------------------------------------------------------------
19
- # In a full app, you would load these from a JSON or CSV text file.
20
  faq_data = [
21
  {
22
  "question": "What is the refund policy for Simplilearn courses?",
@@ -36,7 +35,6 @@ faq_data = [
36
  }
37
  ]
38
 
39
- # Extract text chunks and metadata for the Vector Database
40
  documents = [item["answer"] for item in faq_data]
41
  metadatas = [{"question": item["question"]} for item in faq_data]
42
  ids = [f"faq_{i}" for i in range(len(faq_data))]
@@ -44,10 +42,8 @@ ids = [f"faq_{i}" for i in range(len(faq_data))]
44
  # -------------------------------------------------------------------------
45
  # 2. RAG Pipeline Initialization (CPU-Optimized)
46
  # -------------------------------------------------------------------------
47
- # Initialize a lightweight, highly accurate sentence encoder for local CPU execution
48
  embeddings = HuggingFaceEmbeddings(model_name="all-MiniLM-L6-v2")
49
 
50
- # Create an in-memory Chroma DB Vector Store
51
  vector_store = Chroma.from_texts(
52
  texts=documents,
53
  embedding=embeddings,
@@ -56,8 +52,6 @@ vector_store = Chroma.from_texts(
56
  )
57
  retriever = vector_store.as_retriever(search_kwargs={"k": 2})
58
 
59
- # Setup an Encoder-Decoder generative pipeline via Hugging Face Hosted Inference API
60
- # Mistral-7B-Instruct handles complex reasoning over text efficiently without running locally
61
  llm = HuggingFaceEndpoint(
62
  repo_id="mistralai/Mistral-7B-Instruct-v0.3",
63
  temperature=0.3,
@@ -65,7 +59,6 @@ llm = HuggingFaceEndpoint(
65
  huggingfacehub_api_token=os.getenv("HUGGINGFACEHUB_API_TOKEN")
66
  )
67
 
68
- # Explicitly craft a strict contextual RAG prompt template
69
  prompt_template = """You are Lane, an intelligent, helpful, and friendly Agentic AI Assistant for Simplilearn MOOC.
70
  Use the following pieces of retrieved context to answer the user's question accurately.
71
  If you do not know the answer based on the context, politely state that you do not know.
@@ -79,7 +72,6 @@ Helpful Answer:"""
79
 
80
  rag_prompt = PromptTemplate.from_template(prompt_template)
81
 
82
- # Build the complete LangChain Expression Language (LCEL) chain
83
  def format_docs(docs):
84
  return "\n\n".join(doc.page_content for doc in docs)
85
 
@@ -94,32 +86,28 @@ rag_chain = (
94
  # 3. Application Core Logic
95
  # -------------------------------------------------------------------------
96
  def get_current_time_string():
97
- """Generates the localized timestamp format highlighted in the UI reference."""
98
  now = datetime.datetime.now()
99
  return now.strftime("%dth %B %Y | Time: %I:%M%p").lower()
100
 
101
  def chat_interface(user_message, history):
102
- """Processes inbound user chat messages through the RAG framework."""
103
  if not user_message.strip():
104
  return "", history
105
 
106
- # Process query through our tokenized encoder-decoder RAG architecture
107
  ai_response = rag_chain.invoke(user_message)
108
 
109
- # Update Gradio Chat history payload format
110
  history.append({"role": "user", "content": user_message})
111
  history.append({"role": "assistant", "content": ai_response})
112
 
113
  return "", history
114
 
115
  def handle_quick_reply(button_text, history):
116
- """Bypasses standard empty chat entries to process pre-baked suggestions instantly."""
117
  return chat_interface(button_text, history)
118
 
119
  # -------------------------------------------------------------------------
120
  # 4. Gradio Web Interface Layout
121
  # -------------------------------------------------------------------------
122
- # Injecting custom CSS rules to strictly align font colors, themes, and container styles to your visual specs
123
  custom_css = """
124
  .header-container { background-color: #f7f9fc; padding: 20px; border-radius: 8px; margin-bottom: 15px; }
125
  .meta-container { text-align: right; font-family: 'Georgia', serif; font-style: italic; color: #3b5998; }
@@ -128,9 +116,10 @@ custom_css = """
128
  .quick-btn { background-color: #e0f2f1 !important; border: 1px solid #b2dfdb !important; color: #004d40 !important; font-weight: 600; }
129
  """
130
 
131
- with gr.Blocks(css=custom_css, title="Simplilearn MOOC FAQ Agent") as demo:
 
132
 
133
- # TOP SECTION: Header banner split into Info Branding and Dynamic Meta Controls
134
  with gr.Row(elem_classes=["header-container"]):
135
  with gr.Column(scale=3):
136
  gr.Markdown("""
@@ -140,21 +129,18 @@ with gr.Blocks(css=custom_css, title="Simplilearn MOOC FAQ Agent") as demo:
140
  learners unlock the full potential of specialized MOOC educational pathways.
141
  """)
142
  with gr.Column(scale=2, elem_classes=["meta-container"]):
143
- # UI display showcasing timestamping and system state parameters
144
  time_display = gr.Markdown(value=f"### {get_current_time_string()}")
145
  gr.Markdown("### ☀️ 🌙 ➔ Change Mode (Day / Night)")
146
 
147
- # MAIN SECTION: Split layout dividing Interactive Workspace from Identity Card
148
  with gr.Row():
149
- # Chat Canvas Engine
150
  with gr.Column(scale=4):
 
151
  chatbot = gr.Chatbot(
152
  label="Lane AI Conversation",
153
- type="messages",
154
  value=[{"role": "assistant", "content": "Hello! I'm Lane, your friendly AI Agent. How can I help you today?"}]
155
  )
156
 
157
- # Input Area
158
  with gr.Row():
159
  user_input = gr.Textbox(
160
  show_label=False,
@@ -163,14 +149,12 @@ with gr.Blocks(css=custom_css, title="Simplilearn MOOC FAQ Agent") as demo:
163
  )
164
  submit_btn = gr.Button("Send", variant="primary", scale=1)
165
 
166
- # Quick Reply Interaction Layer
167
  with gr.Row():
168
  btn_refund = gr.Button("Provide refund policy details", elem_classes=["quick-btn"])
169
  btn_cert = gr.Button("Learn about course certification", elem_classes=["quick-btn"])
170
 
171
- # Identity Profile Display Card
172
  with gr.Column(scale=2, elem_classes=["sidebar-panel"]):
173
- # Placeholder for Avatar matching the style from the mockup
174
  gr.HTML("""
175
  <div style='margin-top: 20px;'>
176
  <img src='https://cdn-icons-png.flaticon.com/512/4712/4712010.png' alt='Lane Avatar' style='width:130px; height:130px; object-fit:cover; margin-bottom:15px; border: 4px solid #ffffff;'>
@@ -183,17 +167,14 @@ with gr.Blocks(css=custom_css, title="Simplilearn MOOC FAQ Agent") as demo:
183
  </div>
184
  """)
185
 
186
- # -------------------------------------------------------------------------
187
- # 5. Event Binding Pipeline Connections
188
- # -------------------------------------------------------------------------
189
- # Interactive workflow triggers for text submission
190
  submit_btn.click(chat_interface, inputs=[user_input, chatbot], outputs=[user_input, chatbot])
191
  user_input.submit(chat_interface, inputs=[user_input, chatbot], outputs=[user_input, chatbot])
192
 
193
- # Event routing configuration mapping pre-configured quick buttons to context generation
194
  btn_refund.click(handle_quick_reply, inputs=[btn_refund, chatbot], outputs=[user_input, chatbot])
195
  btn_cert.click(handle_quick_reply, inputs=[btn_cert, chatbot], outputs=[user_input, chatbot])
196
 
197
- # Run production runtime thread configuration
198
  if __name__ == "__main__":
199
- demo.launch(server_name="0.0.0.0", server_port=7860)
 
 
16
  # -------------------------------------------------------------------------
17
  # 1. Mock Data Setup (Simplilearn MOOC FAQs)
18
  # -------------------------------------------------------------------------
 
19
  faq_data = [
20
  {
21
  "question": "What is the refund policy for Simplilearn courses?",
 
35
  }
36
  ]
37
 
 
38
  documents = [item["answer"] for item in faq_data]
39
  metadatas = [{"question": item["question"]} for item in faq_data]
40
  ids = [f"faq_{i}" for i in range(len(faq_data))]
 
42
  # -------------------------------------------------------------------------
43
  # 2. RAG Pipeline Initialization (CPU-Optimized)
44
  # -------------------------------------------------------------------------
 
45
  embeddings = HuggingFaceEmbeddings(model_name="all-MiniLM-L6-v2")
46
 
 
47
  vector_store = Chroma.from_texts(
48
  texts=documents,
49
  embedding=embeddings,
 
52
  )
53
  retriever = vector_store.as_retriever(search_kwargs={"k": 2})
54
 
 
 
55
  llm = HuggingFaceEndpoint(
56
  repo_id="mistralai/Mistral-7B-Instruct-v0.3",
57
  temperature=0.3,
 
59
  huggingfacehub_api_token=os.getenv("HUGGINGFACEHUB_API_TOKEN")
60
  )
61
 
 
62
  prompt_template = """You are Lane, an intelligent, helpful, and friendly Agentic AI Assistant for Simplilearn MOOC.
63
  Use the following pieces of retrieved context to answer the user's question accurately.
64
  If you do not know the answer based on the context, politely state that you do not know.
 
72
 
73
  rag_prompt = PromptTemplate.from_template(prompt_template)
74
 
 
75
  def format_docs(docs):
76
  return "\n\n".join(doc.page_content for doc in docs)
77
 
 
86
  # 3. Application Core Logic
87
  # -------------------------------------------------------------------------
88
  def get_current_time_string():
 
89
  now = datetime.datetime.now()
90
  return now.strftime("%dth %B %Y | Time: %I:%M%p").lower()
91
 
92
  def chat_interface(user_message, history):
 
93
  if not user_message.strip():
94
  return "", history
95
 
96
+ # Generate RAG response
97
  ai_response = rag_chain.invoke(user_message)
98
 
99
+ # Append structured history dictionary mapping (Gradio 5/6 standard native format)
100
  history.append({"role": "user", "content": user_message})
101
  history.append({"role": "assistant", "content": ai_response})
102
 
103
  return "", history
104
 
105
  def handle_quick_reply(button_text, history):
 
106
  return chat_interface(button_text, history)
107
 
108
  # -------------------------------------------------------------------------
109
  # 4. Gradio Web Interface Layout
110
  # -------------------------------------------------------------------------
 
111
  custom_css = """
112
  .header-container { background-color: #f7f9fc; padding: 20px; border-radius: 8px; margin-bottom: 15px; }
113
  .meta-container { text-align: right; font-family: 'Georgia', serif; font-style: italic; color: #3b5998; }
 
116
  .quick-btn { background-color: #e0f2f1 !important; border: 1px solid #b2dfdb !important; color: #004d40 !important; font-weight: 600; }
117
  """
118
 
119
+ # Fixed: Removed 'css' parameter from constructor to satisfy Gradio 6 specifications
120
+ with gr.Blocks(title="Simplilearn MOOC FAQ Agent") as demo:
121
 
122
+ # TOP SECTION
123
  with gr.Row(elem_classes=["header-container"]):
124
  with gr.Column(scale=3):
125
  gr.Markdown("""
 
129
  learners unlock the full potential of specialized MOOC educational pathways.
130
  """)
131
  with gr.Column(scale=2, elem_classes=["meta-container"]):
 
132
  time_display = gr.Markdown(value=f"### {get_current_time_string()}")
133
  gr.Markdown("### ☀️ 🌙 ➔ Change Mode (Day / Night)")
134
 
135
+ # MAIN ROW SECTION
136
  with gr.Row():
 
137
  with gr.Column(scale=4):
138
+ # Fixed: Removed 'type="messages"' parameter since it's now default
139
  chatbot = gr.Chatbot(
140
  label="Lane AI Conversation",
 
141
  value=[{"role": "assistant", "content": "Hello! I'm Lane, your friendly AI Agent. How can I help you today?"}]
142
  )
143
 
 
144
  with gr.Row():
145
  user_input = gr.Textbox(
146
  show_label=False,
 
149
  )
150
  submit_btn = gr.Button("Send", variant="primary", scale=1)
151
 
 
152
  with gr.Row():
153
  btn_refund = gr.Button("Provide refund policy details", elem_classes=["quick-btn"])
154
  btn_cert = gr.Button("Learn about course certification", elem_classes=["quick-btn"])
155
 
156
+ # RIGHT SIDEBAR PANEL
157
  with gr.Column(scale=2, elem_classes=["sidebar-panel"]):
 
158
  gr.HTML("""
159
  <div style='margin-top: 20px;'>
160
  <img src='https://cdn-icons-png.flaticon.com/512/4712/4712010.png' alt='Lane Avatar' style='width:130px; height:130px; object-fit:cover; margin-bottom:15px; border: 4px solid #ffffff;'>
 
167
  </div>
168
  """)
169
 
170
+ # Event Connections
 
 
 
171
  submit_btn.click(chat_interface, inputs=[user_input, chatbot], outputs=[user_input, chatbot])
172
  user_input.submit(chat_interface, inputs=[user_input, chatbot], outputs=[user_input, chatbot])
173
 
 
174
  btn_refund.click(handle_quick_reply, inputs=[btn_refund, chatbot], outputs=[user_input, chatbot])
175
  btn_cert.click(handle_quick_reply, inputs=[btn_cert, chatbot], outputs=[user_input, chatbot])
176
 
177
+ # Run configuration
178
  if __name__ == "__main__":
179
+ # Fixed: Injected custom_css string here inside the launch method call to silence user warnings
180
+ demo.launch(server_name="0.0.0.0", server_port=7860, css=custom_css)