danbooru2023_index / README.md
narugo's picture
Update README.md
542999d verified
metadata
license: mit
task_categories:
  - image-classification
  - image-to-image
  - text-to-image
language:
  - en
  - ja
size_categories:
  - 1M<n<10M

Tar index files for nyanko7/danbooru2023.

You can download images from both nyanko7/danbooru2023 and deepghs/danbooru_newest with cheesechaser.

from cheesechaser.datapool import DanbooruNewestDataPool

pool = DanbooruNewestDataPool()

# download danbooru original images from 7200000-7201000, to directory /data/danbooru_original
pool.batch_download_to_directory(
    resource_ids=range(7200000, 7201000),
    dst_dir='/data/danbooru_original',
    max_workers=12,
)