app.py

import os
from threading import Thread
from typing import Iterator

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


MAX_INPUT_TOKEN_LENGTH = 8192

DESCRIPTION = """\
# CataLlama-v0.2-Instruct-DPO
This Space demonstrates model [CataLlama-v0.2-Instruct-DPO](https://huggingface.co/catallama/CataLlama-v0.2-Instruct-DPO).

CataLlama is a fine-tune of Llama-3-8B to enhance it's proficiency on the Catalan Language.

The model is capable of performing the following **tasks in Catalan**:

- Translation from English to Catalan and Catalan to English
- Summarization - both short form and long form
- Information extraction (suitable for RAG)
- Named Entity Recognition (NER)
- Open question answering
- Sentiment analysis
"""

LICENSE = """\
As a derivate work of [Meta-Llama-3-8B](https://huggingface.co/meta-llama/Meta-Llama-3-8B) by Meta, this demo is governed by the original [llama-3 license](https://llama.meta.com/llama3/license)
"""

if not torch.cuda.is_available():
    DESCRIPTION += "\n<p>Running on CPU 🥶 This demo does not work on CPU.</p>"


if torch.cuda.is_available():
    model_id = "catallama/CataLlama-v0.2-Instruct-SFT"
    model = AutoModelForCausalLM.from_pretrained(model_id, device_map="auto")
    tokenizer = AutoTokenizer.from_pretrained(model_id)


@spaces.GPU(duration=120)
def generate(
    message: str,
    chat_history: list[tuple[str, str]],
    system_prompt: str,
    max_new_tokens: int,
    temperature: float,
    top_p: float,
) -> Iterator[str]:
    
    conversation = []
    if system_prompt:
        conversation.append({"role": "system", "content": system_prompt})
        
    for user, assistant in chat_history:
        conversation.extend([{"role": "user", "content": user}, {"role": "assistant", "content": assistant}])
        
    conversation.append({"role": "user", "content": message})

    input_ids = tokenizer.apply_chat_template(conversation, return_tensors="pt")
    if input_ids.shape[1] > MAX_INPUT_TOKEN_LENGTH:
        input_ids = input_ids[:, -MAX_INPUT_TOKEN_LENGTH:]
        gr.Warning(f"Trimmed input from conversation as it was longer than {MAX_INPUT_TOKEN_LENGTH} tokens.")
    input_ids = input_ids.to(model.device)

    streamer = TextIteratorStreamer(tokenizer, timeout=10.0, skip_prompt=True, skip_special_tokens=True)
    generate_kwargs = dict(
        {"input_ids": input_ids},
        streamer=streamer,
        max_new_tokens=max_new_tokens,
        do_sample=True,
        top_p=top_p,
        temperature=temperature,
    )
    t = Thread(target=model.generate, kwargs=generate_kwargs)
    t.start()

    outputs = []
    for text in streamer:
        outputs.append(text)
        yield "".join(outputs)



chat_interface = gr.ChatInterface(
    fn=generate,
    additional_inputs=[
        gr.Textbox(
            value="Ets un chatbot amigable. Responeu preguntes i ajudeu els usuaris.", 
            label="System message", 
            lines=6
        ),
        gr.Slider(
            minimum=1, 
            maximum=2048, 
            value=1024, 
            step=256, 
            label="Max new tokens"
        ),
        gr.Slider(
            minimum=0.1, 
            maximum=1.0, 
            value=0.3, 
            step=0.05, 
            label="Temperature"
        ),
        gr.Slider(
            minimum=0.1,
            maximum=1.0,
            value=0.90,
            step=0.05,
            label="Top-p (nucleus sampling)",
        ),
    ],
    examples=[
        ["A quina velocitat poden volar els cocodrils?"],
        ["Explica pas a pas com resoldre l'equació següent: 2x + 10 = 0"],
        ["Pot Donald Trump sopar amb Juli Cèsar?"],
    ],
)

with gr.Blocks(css="style.css") as demo:
    gr.Markdown(DESCRIPTION)
    chat_interface.render()
    gr.Markdown(LICENSE)


if __name__ == "__main__":
    demo.queue(max_size=20).launch()