app.py

import gradio as gr

from peft import PeftModel, PeftConfig
from transformers import (
    MistralForCausalLM,
    TextIteratorStreamer,
    AutoTokenizer,
    BitsAndBytesConfig,
    GenerationConfig,
)
from time import sleep
from threading import Thread
from torch import float16
import spaces
import huggingface_hub
from threading import Thread
from queue import Queue
from time import sleep
from os import getenv
from data_logger import log_data
from datetime import datetime

def check_thread(logging_queue: Queue):
    logging_callback = log_data(
        hf_token=getenv("HF_API_TOKEN"),
        dataset_name=getenv("OUTPUT_DATASET"),
        private=True,
    )
    while True:
        sleep(60)
        batch = []
        while not logging_queue.empty():
            batch.append(logging_queue.get())
        if len(batch) > 0:
            try:
                logging_callback(batch)
            except:
                print(
                    "Error happened while pushing data to HF. Puttting items back in queue..."
                )
                for item in batch:
                    logging_queue.put(item)


if getenv("HF_API_TOKEN") is not None:
    #print("Starting logging thread...")
    #log_queue = Queue()
    #t = Thread(target=check_thread, args=(log_queue,))
    #t.start()
    
    logging_callback = log_data(
        hf_token=getenv("HF_API_TOKEN"),
        dataset_name=getenv("OUTPUT_DATASET"),
        private=True,
    )
else:
    print("No HF_API_TOKEN found. Logging is disabled.")


config = PeftConfig.from_pretrained("lang-uk/dragoman")
quant_config = BitsAndBytesConfig(
    load_in_4bit=True,
    bnb_4bit_quant_type="nf4",
    bnb_4bit_compute_dtype=float16,
    bnb_4bit_use_double_quant=False,
)

model = MistralForCausalLM.from_pretrained(
    "mistralai/Mistral-7B-v0.1", quantization_config=quant_config
)


# device_map="auto",)
model = PeftModel.from_pretrained(model, "lang-uk/dragoman").to("cuda")
tokenizer = AutoTokenizer.from_pretrained(
    "mistralai/Mistral-7B-v0.1", use_fast=False, add_bos_token=False
)


@spaces.GPU(duration=30)
def translate(input_text):
    global log_queue
    # generated_text = ""
    input_text = input_text.strip()
    print(f"{datetime.utcnow()} | Translating: {input_text}")

    if getenv("HF_API_TOKEN") is not None:
        try:
            logging_callback = log_data(
                hf_token=getenv("HF_API_TOKEN"),
                dataset_name=getenv("OUTPUT_DATASET"),
                private=True,
            )
            logging_callback([[input_text]])
        except:
            print("Error happened while pushing data to HF.")

    input_text = f"[INST] {input_text} [/INST]"
    inputs = tokenizer([input_text], return_tensors="pt").to(model.device)

    generation_kwargs = dict(
        inputs, max_new_tokens=200, num_beams=10, temperature=1, pad_token_id=tokenizer.eos_token_id
    )  # streamer=streamer,

    # streaming support
    # streamer = TextIteratorStreamer(
    #    tokenizer, skip_prompt=True, skip_special_tokens=True
    # )

    # thread = Thread(target=model.generate, kwargs=generation_kwargs)

    # thread.start()

    # for new_text in streamer:
    #    generated_text += new_text
    #    yield generated_text

    # generated_text += "\n"
    # yield generated_text

    output = model.generate(**generation_kwargs)
    output = (
        tokenizer.decode(output[0], skip_special_tokens=True)
        .split("[/INST] ")[-1]
        .strip()
    )
    return output


# download description of the model
desc_file = huggingface_hub.hf_hub_download("lang-uk/dragoman", "README.md")

with open(desc_file, "r") as f:
    model_description = f.read()
    model_description = model_description[model_description.find("---", 1) + 5 :]
    model_description = (
        """### By using this service, users are required to agree to the following terms: you agree that user input will be collected for future research and model improvements. \n\n"""
        + model_description
    )


iface = gr.Interface(
    fn=translate,
    inputs=gr.Textbox(
        value='This demo contains a model from paper "Setting up the Data Printer with Improved English to Ukrainian Machine Translation", accepted to UNLP 2024 workshop at the LREC-COLING 2024 conference.',
        label="Source sentence",
    ),
    outputs=gr.Textbox(
        value='Ця демо-версія містить модель із статті "Налаштування принтера даних із покращеним машинним перекладом з англійської на українську", яка була прийнята до семінару UNLP 2024 на конференції LREC-COLING 2024.',
        label="Translated sentence",
    ),
    examples=[
        [
            "The Colosseum in Rome was a symbol of the grandeur and power of the Roman Empire and was a place for the emperor to connect with the people by providing them with entertainment and free food."
        ],
        [
            "How many leaves would it drop in a month of February in a non-leap year?",
        ],
        [
            "ChatGPT (Chat Generative Pre-trained Transformer) is a chatbot developed by OpenAI and launched on November 30, 2022. Based on a large language model, it enables users to refine and steer a conversation towards a desired length, format, style, level of detail, and language. Successive prompts and replies, known as prompt engineering, are considered at each conversation stage as a context.[2] ",
        ],
        [
            "who holds this neighborhood?",
        ],
        
    ],
    title="Dragoman: SOTA English-Ukrainian translation model",
    description='This demo contains a model from paper "Setting up the Data Printer with Improved English to Ukrainian Machine Translation", accepted to UNLP 2024 workshop at the LREC-COLING 2024 conference.',
    article=model_description,
    # thumbnail: str | None = None,
    # css: str | None = None,
    # batch: bool = False,
    # max_batch_size: int = 4,
    # api_name: str | Literal[False] | None = "predict",
    submit_btn="Translate",
)
iface.launch()