Upload app.py

app.py

@@ -108,12 +108,12 @@ def llm_query(user_input,company):
 
     # 2 - Create messages
     prompt = [
-        {'role':'system', 'content': qna_system_message},
-        {'role': 'user', 'content': qna_user_message_template.format(
+        SystemMessage(content=qna_system_message),
+        HumanMessage(content=qna_user_message_template.format(
             context=context_for_query,
             question=user_input
             )
-        }
+        )
     ]
 
     # Get response from the LLM
@@ -155,11 +155,11 @@ def llm_query(user_input,company):
 
 # Set-up the Gradio UI
 textbox = gr.Textbox(label='Question:') # Create a textbox for user input
-company = gr.Radio(label='Company:', choices=["aws", "google", "ibm", "meta", "microsoft"], value="aws") # Create a radio button for company selection
+company = gr.Radio(label='Company:', choices=["aws", "google", "IBM", "Meta", "msft"], value="aws") # Create a radio button for company selection
 
 # Create Gradio interface
 # For the inputs parameter of Interface provide [textbox,company] with outputs parameter of Interface provide prediction
-demo = gr.Interface(fn=llm_query, inputs=[textbox, company], outputs="text", title="Financial Analyst Assistant", description="Ask questions about the financial performance of AWS, Google, IBM, Meta, and Microsoft based on their 10-K reports.\n\nPlease enter a question below.")
+demo = gr.Interface(fn=llm_query, inputs=[textbox, company], outputs="text", title="Financial Analyst Assistant", description="Ask questions about the financial performance of AWS, Google, IBM, Meta, and Microsoft based on their 10-K reports.\n\nPlease enter a question below.", theme=gr.themes.Soft())
 
 demo.queue()
 demo.launch()