Added response function

app.py

@@ -41,5 +41,28 @@ def pull_relevant_info(query):
 
 client = InferenceClient("HuggingFaceH4/zephyr-7b-beta")
 
+def respond(message, history): 
+    
+    info = pull_relevant_info(message)
+    system_message = (f"You are a friendly chatbot. Use the following information to help answer the user's question:\n{info}\n")
+    messages = [{"role": "system", "content": system_message}]
+    
+    if history:
+        messages.extend(history)
+        
+    messages.append({"role": "user", "content": message})
+
+    response = ""
+    
+    for message_chunk in client.chat_completion(
+        messages,
+        max_tokens=100,
+        stream=True
+    ):
+        token = message_chunk['choices'][0]['delta'].get('content', '')
+        respond += token
+
+    yield response
+    
 chatbot.launch()