import torch    
from transformers import AutoTokenizer,AutoModelForMaskedLM, GPT2LMHeadModel,GPT2Tokenizer, AutoModelForSeq2SeqLM, AutoModelForCausalLM
tokenizer = GPT2Tokenizer.from_pretrained("microsoft/DialoGPT-medium")

model = GPT2LMHeadModel.from_pretrained('Stage v3.0')
# Let's chat for 4 lines
                                                    
for step in range(50):
    # encode the new user input, add the eos_token and return a tensor in Pytorch
    new_user_input_ids = tokenizer.encode(input(">> You:") + tokenizer.eos_token, return_tensors='pt')
    # print(new_user_input_ids)


    # append the new user input tokens to the chat history
    bot_input_ids = torch.cat([ new_user_input_ids], dim=-1) if step > 0 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=200,
        pad_token_id=tokenizer.eos_token_id,  
        no_repeat_ngram_size=3,       
        do_sample=True, 
        top_k=100, # It controls the diversity of the generated output; the model considers the top 100 tokens
        top_p=0.9,# tokens with a cumulative probability higher than 0.9 are excluded.  
        temperature=0.9 # It controls the randomness of the generated output
    )
    
    # pretty print last ouput tokens from bot
    print("Chatbot: {}".format(tokenizer.decode(chat_history_ids[:, bot_input_ids.shape[-1]:][0], skip_special_tokens=True)))