natural_instructions

  • 説明:

説明

自然な指示として表現された 1600 以上のタスクのコンパイル。元のタスク コレクションは、 https ://github.com/allenai/natural-instructions にあります。この元のバージョンには前処理や変更は加えられていません。

このタスク コレクションのユーザーは、含まれているデータセットの基礎となるライセンスを参照し、それに応じて引用する必要があることに注意してください。

スプリット
'train' 5,040,134
  • 機能構造:
FeaturesDict({
    'definition': Text(shape=(), dtype=string),
    'id': Text(shape=(), dtype=string),
    'input': Text(shape=(), dtype=string),
    'output': Text(shape=(), dtype=string),
    'source': Text(shape=(), dtype=string),
    'task_name': Text(shape=(), dtype=string),
})
  • 機能のドキュメント:
特徴クラスDtype説明
特徴辞書
意味文章ストリング
ID文章ストリング
入力文章ストリング
出力文章ストリング
ソース文章ストリング
タスク名文章ストリング
  • 引用
@article{wang2022benchmarking,
  title={Benchmarking generalization via in-context instructions on 1,600+ language tasks},
  author={Wang, Yizhong and Mishra, Swaroop and Alipoormolabashi, Pegah and Kordi, Yeganeh and Mirzaei, Amirreza and Arunkumar, Anjana and Ashok, Arjun and Dhanasekaran, Arut Selvan and Naik, Atharva and Stap, David and others},
  journal={arXiv preprint arXiv:2204.07705},
  year={2022}
}
,

  • 説明:

説明

自然な指示として表現された 1600 以上のタスクのコンパイル。元のタスク コレクションは、 https ://github.com/allenai/natural-instructions にあります。この元のバージョンには前処理や変更は加えられていません。

このタスク コレクションのユーザーは、含まれているデータセットの基礎となるライセンスを参照し、それに応じて引用する必要があることに注意してください。

スプリット
'train' 5,040,134
  • 機能構造:
FeaturesDict({
    'definition': Text(shape=(), dtype=string),
    'id': Text(shape=(), dtype=string),
    'input': Text(shape=(), dtype=string),
    'output': Text(shape=(), dtype=string),
    'source': Text(shape=(), dtype=string),
    'task_name': Text(shape=(), dtype=string),
})
  • 機能のドキュメント:
特徴クラスDtype説明
特徴辞書
意味文章ストリング
ID文章ストリング
入力文章ストリング
出力文章ストリング
ソース文章ストリング
タスク名文章ストリング
  • 引用
@article{wang2022benchmarking,
  title={Benchmarking generalization via in-context instructions on 1,600+ language tasks},
  author={Wang, Yizhong and Mishra, Swaroop and Alipoormolabashi, Pegah and Kordi, Yeganeh and Mirzaei, Amirreza and Arunkumar, Anjana and Ashok, Arjun and Dhanasekaran, Arut Selvan and Naik, Atharva and Stap, David and others},
  journal={arXiv preprint arXiv:2204.07705},
  year={2022}
}