examples/add_punctuation/download_model.py

#!/usr/bin/python3
# -*- coding: utf-8 -*-
import argparse
import os
from pathlib import Path
import sys

pwd = os.path.abspath(os.path.dirname(__file__))
sys.path.append(os.path.join(pwd, "../../"))

import huggingface_hub

from project_settings import project_path


def get_args():
    parser = argparse.ArgumentParser()

    parser.add_argument(
        "--repo_id",
        default="csukuangfj/sherpa-onnx-punct-ct-transformer-zh-en-vocab272727-2024-04-12",
        type=str
    )
    parser.add_argument("--model_filename", default="model.onnx", type=str)
    parser.add_argument("--model_sub_folder", default=".", type=str)

    parser.add_argument(
        "--pretrained_model_dir",
        default=(project_path / "pretrained_models").as_posix(),
        type=str
    )
    args = parser.parse_args()
    return args


def main():
    args = get_args()

    pretrained_model_dir = Path(args.pretrained_model_dir)
    pretrained_model_dir.mkdir(exist_ok=True)

    repo_id: Path = Path(args.repo_id)
    local_model_dir = pretrained_model_dir / "huggingface" / repo_id
    local_model_dir.mkdir(parents=True, exist_ok=True)

    print("download model")
    model_filename = huggingface_hub.hf_hub_download(
        repo_id=args.repo_id,
        filename=args.model_filename,
        subfolder=args.model_sub_folder,
        local_dir=local_model_dir.as_posix(),
    )
    print(model_filename)
    return


if __name__ == "__main__":
    main()