metadata
dataset_info:
features:
- name: type
dtype: string
- name: id
dtype: int64
- name: title
dtype: string
- name: other_names
sequence: string
splits:
- name: train
num_bytes: 2500556
num_examples: 33658
download_size: 1413095
dataset_size: 2500556
configs:
- config_name: default
data_files:
- split: train
path: data/train-*
isek-ai/danbooru-wiki-2024 のdanbooruタグと日本語タグとの対応を Calm3 を使ってフィルタリング、補完したもの (仮)