Update app.py

app.py

@@ -54,11 +54,19 @@ def stream_chat(
 ):
     print(f'message: {message}')
     print(f'history: {history}')
+
+    conversation = []
+    for prompt, answer in history:
+        conversation.extend([
+            {"role": "user", "content": prompt}, 
+            {"role": "assistant", "content": answer},
+        ])
+    
     torch.manual_seed(0)
     resp, history = model.chat(
         tokenizer,
         query = message,
-        history = history,
+        history = conversation,
         max_length = max_new_tokens,
         do_sample = False if temperature == 0 else True,
         top_p = top_p,