Update app.py

app.py

@@ -34,7 +34,7 @@ documents = SimpleDirectoryReader(
 ).load_data()
 
 index = VectorStoreIndex.from_documents(documents)
-
+ft_model_id='ft:gpt-3.5-turbo-0613:handshake::7raTNuPU'
 
 @cl.on_chat_start
 async def factory():
@@ -59,31 +59,51 @@ async def factory():
      callback_manager=CallbackManager([cl.LlamaIndexCallbackHandler()]),
     )
 
+    ft_context = ServiceContext.from_defaults(
+    llm=OpenAI(model=ft_model_id, temperature=0.3),
+    context_window=2048,  # limit the context window artifically to test refine process
+  callback_manager=CallbackManager([cl.LlamaIndexCallbackHandler()]),
+)
+
     query_engine = index.as_query_engine(
         service_context=gpt_35_context
     )
 
+    finetune_query_engine = index.as_query_engine(service_context=ft_context)
+
     msg.content = f"Index built!"
     await msg.send()
 
     cl.user_session.set("query_engine", query_engine)
 
+    cl.user_session.set("finetune_query_engine", finetune_query_engine)
+
 
 @cl.on_message
 async def main(message):
     query_engine = cl.user_session.get("query_engine")  # type: RetrieverQueryEngine
     response = await cl.make_async(query_engine.query)(message)
-    print(response)
-    
-    # await cl.Message(content=response).send()
+
     response_message = cl.Message(content="")
-    response_message.content = response
-    print(response_message)
+    response_message.content = "Base Model Response"
+
+    elements = [
+    cl.Text(content=response.response, display="inline")
+]
+    response_message.elements = elements
+
     await response_message.send()
-    # # for token in response.response_gen:
-    # #     await response_message.stream_token(token=token)
 
-    # # if response.response_txt:
-    # response_message.content = response
+    finetune_query_engine = cl.user_session.get("finetune_query_engine")  # type: RetrieverQueryEngine
+    finetune_response = await cl.make_async(finetune_query_engine.query)(message)
+    
+    finetune_response_message = cl.Message(content="")
+    finetune_response_message.content = 'Finetune Model Response'
+
+    elements = [
+    cl.Text( content=finetune_response.response, display="inline")
+]
+    finetune_response_message.elements = elements
+
+    await finetune_response_message.send()
 
-    # await response_message.send()