Update app.py

app.py

@@ -179,6 +179,7 @@ def create_vector_db(final_items):
                 top_p=0.9, 
                 top_k=20, 
                 stream=False,
+                verbose = True
             )
     
             # Extract the generated summary text
@@ -241,21 +242,22 @@ def generate_response(db, query_text, previous_context):
         stream=True,  # Enable streaming
         temperature=0.1,
         top_p=0.9,
-        top_k=20
+        top_k=20,
+        verbose = True
     )
     
     # Use Streamlit to stream the response in real-time
-    temp_response = ""
     full_response = ""
     
     response_placeholder = st.empty()  # Create a placeholder for streaming output
+    
+    # Stream the response token by token
     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
+        full_response += token_text
         
-        # Update the placeholder in real-time with the new token
-        response_placeholder.write(temp_response)  
+        # Continuously update the placeholder in real-time with the new token
+        response_placeholder.write(full_response)  
     
     return full_response