Update app.py

app.py

@@ -48,38 +48,37 @@ llm_name = "gpt-3.5-turbo"
 
 vectordb = initialize.initialize()
 
-def chat_query(question):
+def chat_query(question, history):
     
     llm = ChatOpenAI(model=llm_name, temperature=0.1, api_key = OPENAI_API_KEY)
 
     # Conversation Retrival Chain with Memory
     memory = ConversationBufferMemory(memory_key="chat_history", return_messages=True)
     retriever=vectordb.as_retriever()
-    #qa = ConversationalRetrievalChain.from_llm(llm, retriever=retriever, memory=memory)
+    qa = ConversationalRetrievalChain.from_llm(llm, retriever=retriever, memory=memory)
 
     # Replace input() with question variable for Gradio
-    # result = qa({"question": question})
-    # return result['answer']
+    result = qa({"question": question})
+    return result['answer']
 
     # Chatbot only answers based on Documents
-    qa = VectorDBQA.from_chain_type(llm=OpenAI(openai_api_key = OPENAI_API_KEY, ), chain_type="stuff", vectorstore=vectordb)
-    result = qa.run(question)
-    return result
+    # qa = VectorDBQA.from_chain_type(llm=OpenAI(openai_api_key = OPENAI_API_KEY, ), chain_type="stuff", vectorstore=vectordb)
+    # result = qa.run(question)
+    # return result
 
     
 
 
 # logo_path = os.path.join(os.getcwd(), "Logo.png")
 
-iface = gr.Interface(
+iface = gr.ChatInterface(
     fn=chat_query,
-    inputs= gr.Textbox(lines = 6, placeholder="Enter your Query here....",label="Query :"),
-    outputs=gr.Textbox(label="Chatbot Reply : "),
     title  = " -----:  ChatBot  :----- ",
     description="""-- Welcome to the Language Model trained on Model-TS (Engineering-SS).\n\n
                    -- The Model tries to answer the Query based on Model-Technical Specifications. \n\n
                    -- For precise reply, please input `Specific Keywords` in your Query. \n\n """,
     concurrency_limit = None,
+    examples = {'What should be the GIB height outside the GIS hall ?'},
     
 )