Update app.py

app.py

@@ -181,7 +181,6 @@ def create_vector_db(final_items):
     
     return db
 
-
 def generate_response(db, query_text, previous_context):
     query_results = db.query(
         query_texts=query_text,
@@ -230,12 +229,18 @@ def generate_response(db, query_text, previous_context):
     
     # Use Streamlit to stream the response in real-time
     temp_response = ""
+    full_response = ""
+    
+    response_placeholder = st.empty()  # Create a placeholder for streaming output
     for token in output_stream:
         token_text = token["choices"][0]["text"]
+        full_response += token_text  # Keep the entire response for future context
         temp_response += token_text
-        st.write(temp_response)  # Update the Streamlit UI with the current response
-
-    return temp_response
+        
+        # Update the placeholder in real-time with the new token
+        response_placeholder.write(temp_response)  
+    
+    return full_response
 
 def streamlit_app():
     st.title("BioModelsRAG")
@@ -295,3 +300,4 @@ def streamlit_app():
 
 if __name__ == "__main__":
     streamlit_app()
+