Datasets:
metadata
annotations_creators:
- crowdsourced
language_creators:
- crowdsourced
language:
- zh
license:
- cc-by-4.0
multilinguality:
- monolingual
pretty_name: laion2B-multi-chinese-subset
size_categories:
- 10M<n<100M
task_categories:
- feature-extraction
laion2B-multi-chinese-subset
Wanng: 正在更新。预计国庆后正式上线。
预计一共143M个中文图文对。
Dataset Description
- Homepage: laion-5b
- Huggingface: laion/laion2B-multi
Lisence
CC-BY-4.0