Update app.py

Train model before generate text

app.py

@@ -23,6 +23,8 @@ faq_embeddings = embedding_model.encode(faq_questions, convert_to_numpy=True)
 index = faiss.IndexFlatL2(faq_embeddings.shape[1])
 index.add(faq_embeddings)
 
+
+
 def retrieve_uber_info(query):
     """Retrieve the most relevant Uber FAQ answer for the given query."""
     query_embedding = embedding_model.encode([query], convert_to_numpy=True)
@@ -35,35 +37,27 @@ def retrieve_uber_info(query):
 client = InferenceClient("HuggingFaceH4/zephyr-7b-beta")
 
 
-def respond(
-    message,
-    history: list[tuple[str, str]],
-    system_message,
-    max_tokens,
-    temperature,
-    top_p,
-):
-    messages = [{"role": "system", "content": system_message}]
+def respond(message, history, system_message, max_tokens, temperature, top_p):
+    """Generate a response using Zephyr 7B while integrating retrieved Uber knowledge."""
+    
+    retrieved_answer = retrieve_uber_info(message)
+    system_message += f"\n\nUber FAQ Context: {retrieved_answer}"
 
+    messages = [{"role": "system", "content": system_message}]
+    
     for val in history:
         if val[0]:
             messages.append({"role": "user", "content": val[0]})
         if val[1]:
             messages.append({"role": "assistant", "content": val[1]})
-
+    
     messages.append({"role": "user", "content": message})
 
     response = ""
-
     for message in client.chat_completion(
-        messages,
-        max_tokens=max_tokens,
-        stream=True,
-        temperature=temperature,
-        top_p=top_p,
+        messages, max_tokens=max_tokens, stream=True, temperature=temperature, top_p=top_p
     ):
         token = message.choices[0].delta.content
-
         response += token
         yield response