import os
import gradio as gr

HF_TOKEN = os.getenv('HF_TOKEN')
hf_writer = gr.HuggingFaceDatasetSaver(HF_TOKEN, "Rick-bot-flags")

title = "Ask Rick a Question"
description = """
<center>
The bot was trained to answer questions based on Rick and Morty dialogues. Ask Rick anything!
![](rick.png)
</center>
"""
article = "Check out (the original Rick and Morty Bot)[https://huggingface.co/spaces/kingabzpro/Rick_and_Morty_Bot] that this demo is based off of."

from transformers import AutoModelForCausalLM, AutoTokenizer
import torch

tokenizer = AutoTokenizer.from_pretrained("ericzhou/DialoGPT-Medium-Rick_v2")
model = AutoModelForCausalLM.from_pretrained("ericzhou/DialoGPT-Medium-Rick_v2")

def predict(input):
    # tokenize the new input sentence
    new_user_input_ids = tokenizer.encode(input + tokenizer.eos_token, return_tensors='pt')

    # append the new user input tokens to the chat history
    bot_input_ids = torch.cat([torch.LongTensor(history), new_user_input_ids], dim=-1)

    # generate a response 
    history = model.generate(bot_input_ids, max_length=1000, pad_token_id=tokenizer.eos_token_id).tolist()

    # convert the tokens to text, and then split the responses into the right format
    response = tokenizer.decode(history[0]).split("<|endoftext|>")
    return response[1]

gr.Interface(fn = predict, inputs = ["textbox"], outputs = ["text"],allow_flagging = "manual",title = title, flagging_callback = hf_writer, description = description, article = article ).launch(enable_queue=True)  # customizes the input component