Update app.py

app.py

@@ -91,12 +91,17 @@ def predict(inputs, top_p, temperature, chat_counter, chatbot, history, request:
     try:
         # make a POST request to the API endpoint using the requests.post method, passing in stream=True
         response = requests.post(API_URL, headers=headers, json=payload, stream=True)
+        print ('r', response)
+        import sys
+        sys.stdout.flush()
         response_code = f"{response}"
         #if response_code.strip() != "<Response [200]>":
         #    #print(f"response code - {response}")
         #    raise Exception(f"Sorry, hitting rate limit. Please try again later. {response}")
         
         for chunk in response.iter_lines():
+            print (chunk)
+            sys.stdout.flush()
             #Skipping first chunk
             if counter == 0:
                 counter += 1