from transformers import AutoModelForCausalLM, AutoTokenizer    
import gradio as gr    
import torch    
  
tokenizer = AutoTokenizer.from_pretrained("microsoft/DialoGPT-medium")  
tokenizer.padding_side = 'left'    
model = AutoModelForCausalLM.from_pretrained("microsoft/DialoGPT-medium")   
  
class ChatBot:    
    def __init__(self):    
        self.history = []    
  
    def predict(self, input):        
        new_user_input_ids = tokenizer.encode(input + tokenizer.eos_token, return_tensors="pt")        
        flat_history = [item for sublist in self.history for item in sublist]      
        bot_input_ids = torch.cat([torch.tensor(flat_history), new_user_input_ids], dim=-1) if self.history else new_user_input_ids        
        chat_history_ids = model.generate(bot_input_ids, max_length=2000, pad_token_id=tokenizer.eos_token_id)        
        self.history.append(chat_history_ids[:, bot_input_ids.shape[-1]:].tolist()[0])        
        response = tokenizer.decode(chat_history_ids[:, bot_input_ids.shape[-1]:][0], skip_special_tokens=True)        
        return response  
  
bot = ChatBot()    
  
title = "👋🏻Welcome to Tonic's EZ Chat🚀"    
description = "You can use this Space to test out the current model (DialoGPT-medium) or duplicate this Space and use it for any other model on 🤗HuggingFace."    
examples = [["How are you?"]]    
  
iface = gr.Interface(    
    fn=bot.predict,    
    title=title,    
    description=description,    
    examples=examples,    
    inputs="text",    
    outputs="text", 
    theme="ParityError/Anime"
)    
  
iface.launch()