|
|
|
|
|
|
|
|
import os |
|
|
from collections.abc import Iterator |
|
|
from threading import Thread |
|
|
|
|
|
import gradio as gr |
|
|
import spaces |
|
|
import torch |
|
|
from transformers import AutoModelForCausalLM, AutoTokenizer, TextIteratorStreamer |
|
|
|
|
|
DESCRIPTION = " " |
|
|
|
|
|
if not torch.cuda.is_available(): |
|
|
DESCRIPTION += "\n<p>Running on CPU 🥶 This demo does not work on CPU.</p>" |
|
|
|
|
|
MAX_INPUT_TOKEN_LENGTH = int(os.getenv("MAX_INPUT_TOKEN_LENGTH", "4096")) |
|
|
|
|
|
if torch.cuda.is_available(): |
|
|
model_id = "mistralai/Mistral-7B-Instruct-v0.3" |
|
|
model = AutoModelForCausalLM.from_pretrained(model_id, torch_dtype=torch.float16, device_map="auto") |
|
|
tokenizer = AutoTokenizer.from_pretrained(model_id) |
|
|
|
|
|
|
|
|
@spaces.GPU |
|
|
def generate(message: str, chat_history: list[dict]) -> Iterator[str]: |
|
|
"""Fonction de génération sans sliders : les paramètres |
|
|
de génération (max_new_tokens, température, etc.) sont |
|
|
fixés en dur. |
|
|
""" |
|
|
|
|
|
max_new_tokens = 1024 |
|
|
temperature = 0.6 |
|
|
top_p = 0.9 |
|
|
top_k = 50 |
|
|
repetition_penalty = 1.2 |
|
|
|
|
|
|
|
|
conversation = [*chat_history, {"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=20.0, 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, |
|
|
top_k=top_k, |
|
|
temperature=temperature, |
|
|
num_beams=1, |
|
|
repetition_penalty=repetition_penalty, |
|
|
) |
|
|
t = Thread(target=model.generate, kwargs=generate_kwargs) |
|
|
t.start() |
|
|
|
|
|
outputs = [] |
|
|
for text in streamer: |
|
|
outputs.append(text) |
|
|
yield "".join(outputs) |
|
|
|
|
|
|
|
|
|
|
|
demo = gr.ChatInterface( |
|
|
fn=generate, |
|
|
stop_btn=None, |
|
|
examples=[ |
|
|
["C’est quoi le consentement ? Comment savoir si ma copine a envie de moi ?"], |
|
|
["C’est quoi une agression sexuelle ?"], |
|
|
["C’est quoi un viol ?"], |
|
|
["C’est quoi un attouchement ?"], |
|
|
["C’est quoi un harcèlement sexuel ?"], |
|
|
["Est-ce illégal de visionner du porno ?"], |
|
|
["C’est quoi un harcèlement sexuel ?"], |
|
|
["Mon copain me demande un nude, dois-je le faire ?"], |
|
|
["Mon ancien copain me menace de poster des photos de moi nue sur internet, que faire ?"], |
|
|
["Que puis-je faire si un membre de ma famille me touche d’une manière bizarre, mais que j’ai peur de parler ou de ne pas être cru ?" |
|
|
], |
|
|
], |
|
|
type="messages", |
|
|
description=DESCRIPTION, |
|
|
css_paths="style.css", |
|
|
) |
|
|
|
|
|
if __name__ == "__main__": |
|
|
demo.queue(max_size=20).launch(share=True) |
|
|
|
