app.py

from transformers import AutoModelForCausalLM, AutoTokenizer
import gradio as gr
import torch

title = "🤖AI ChatBot"
description = "A State-of-the-Art Large-scale Pretrained Response generation model (gpt-neo-1.3B)"
examples = [["How are you?"]]

# Use the better model and tokenizer
model_name = "EleutherAI/gpt-neo-1.3B"
tokenizer = AutoTokenizer.from_pretrained(model_name)
model = AutoModelForCausalLM.from_pretrained(model_name)

def predict(input_text, history=None):
    if history is None:
        history = []

    # Tokenize the new input sentence
    new_user_input_ids = tokenizer.encode(
        input_text + 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 using batch processing
    generated_ids = model.generate(
        bot_input_ids, max_length=4000, pad_token_id=tokenizer.eos_token_id
    )

    # Convert the generated response tokens to text
    response = tokenizer.decode(generated_ids[0], skip_special_tokens=True)

    # Split the responses into lines
    response = response.split("\n")

    # Convert to tuples of list
    response = [(response[i], response[i + 1]) for i in range(0, len(response) - 1, 2)]

    return response, generated_ids.tolist()

gr.Interface(
    fn=predict,
    title=title,
    description=description,
    examples=examples,
    inputs=["text", "state"],
    outputs=["chatbot", "state"],
    theme="finlaymacklon/boxy_violet",
).launch()