Update app.py

app.py

@@ -3,7 +3,7 @@ import gradio as gr
 from transformers import AutoModelForCausalLM, AutoTokenizer
 from util_funcs import get_length_param
 
-def chat_function(Message, Length_of_the_answer, Who_is_next, Base_to_On_subject):   # model, tokenizer
+def chat_function(Message, Length_of_the_answer, Who_is_next, Base_to_On_subject, history):   # model, tokenizer
     
     input_user = Message
     
@@ -21,9 +21,7 @@ def chat_function(Message, Length_of_the_answer, Who_is_next, Base_to_On_subject
     elif Who_is_next == 'Me':
         next_who = 'H'
         
-        
-    
-    history = gr.get_state() or []
+    history = history or []
     chat_history_ids = torch.zeros((1, 0), dtype=torch.int) if history == [] else torch.tensor(history[-1][2], dtype=torch.long)
 
     # encode the new user input, add parameters and return a tensor in Pytorch
@@ -106,7 +104,8 @@ iface = gr.Interface(chat_function,
                          "text",
                          gr.inputs.Radio(["short", "medium", "long"], default='medium'),
                          gr.inputs.Radio(["Kirill", "Me"], default='Kirill'),
-                         gr.inputs.Slider(0, 1, default=0.5)
+                         gr.inputs.Slider(0, 1, default=0.5),
+                         "state"
                      ],
                      "html",
                      title=title, description=description, article=article, examples=examples,