simplify things

app.py

@@ -10,25 +10,14 @@ import gradio as gr
 def chat(message, history):
     history = history if history is not None else []
     new_user_input_ids = tokenizer.encode(message+tokenizer.eos_token, return_tensors='pt')
-    # append the new user input tokens to the chat history
-    f1 = lambda x: x[2]
-    f2 = lambda x: x[3]
-    if len(history) > 0:
-        bot_input_ids = torch.IntTensor([f(item) for item in history for f in (f1, f2)])
-    else:
-        bot_input_ids = torch.IntTensor([[]])
-    print("bot_input_ids", bot_input_ids)
-    print("new_user_input_ids", new_user_input_ids)
-    print("sizr bot_input_ids", bot_input_ids.size())
-    print("size new_user_input_id", new_user_input_ids.size())
-    bot_input_ids = torch.cat([bot_input_ids, new_user_input_ids], dim=0)  if bot_input_ids is not None else new_user_input_ids
-    # generated a response while limiting the total chat history to 1000 tokens, 
-    chat_history_ids = model.generate(bot_input_ids, max_length=5000, pad_token_id=tokenizer.eos_token_id)
+    bot_input_ids = torch.cat([torch.LongTensor(history), new_user_input_ids], dim=-1)
+    history = model.generate(bot_input_ids, max_length=500, pad_token_id=tokenizer.eos_token_id).tolist()
+    response = tokenizer.decode(history[0]).replace("<|endoftext|>", "\n")
     # pretty print last ouput tokens from bot
-    response =  tokenizer.decode(chat_history_ids[:, bot_input_ids.shape[-1]:][0], skip_special_tokens=True)
+    # response =  tokenizer.decode(chat_history_ids[:, bot_input_ids.shape[-1]:][0], skip_special_tokens=True)
     print("The response is ", [response])
-    history.append((message, response, new_user_input_ids, chat_history_ids))
-    return history, bot_input_ids, feedback(message)
+    # history.append((message, response, new_user_input_ids, chat_history_ids))
+    return response, feedback(message)
 
 
 def feedback(text):