- Descriptif :
Description
Une compilation de plus de 1600 tâches formulées sous forme d'instructions naturelles. La collection de tâches d'origine peut être trouvée sur : https://github.com/allenai/natural-instructions Aucun prétraitement ou modification n'a été apporté à cette version d'origine.
Notez que les utilisateurs de cette collection de tâches doivent consulter les licences sous-jacentes des ensembles de données contenus et les citer en conséquence.
Page d' accueil : https://github.com/allenai/natural-instructions
Code source :
tfds.datasets.natural_instructions.BuilderVersions :
-
1.0.0: Version initiale. -
1.0.1(par défaut) : champ de nom de tâche ajouté et ID fixe utilisé pour le brassage afin d'utiliser des ID stables.
-
Taille du téléchargement :
3.08 GiBTaille du jeu de données :
4.73 GiBMise en cache automatique ( documentation ): Non
Fractionnements :
| Diviser | Exemples |
|---|---|
'train' | 5 040 134 |
- Structure des fonctionnalités :
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),
})
- Documentation des fonctionnalités :
| Caractéristique | Classe | Façonner | Dtype | Description |
|---|---|---|---|---|
| FonctionnalitésDict | ||||
| définition | Texte | chaîne de caractères | ||
| identifiant | Texte | chaîne de caractères | ||
| contribution | Texte | chaîne de caractères | ||
| production | Texte | chaîne de caractères | ||
| la source | Texte | chaîne de caractères | ||
| Nom de la tâche | Texte | chaîne de caractères |
Clés supervisées (Voir
as_superviseddoc ):NoneFigure ( tfds.show_examples ) : non pris en charge.
Exemples ( tfds.as_dataframe ):
- Citation :
@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}
}
, - Descriptif :
Description
Une compilation de plus de 1600 tâches formulées sous forme d'instructions naturelles. La collection de tâches d'origine peut être trouvée sur : https://github.com/allenai/natural-instructions Aucun prétraitement ou modification n'a été apporté à cette version d'origine.
Notez que les utilisateurs de cette collection de tâches doivent consulter les licences sous-jacentes des ensembles de données contenus et les citer en conséquence.
Page d' accueil : https://github.com/allenai/natural-instructions
Code source :
tfds.datasets.natural_instructions.BuilderVersions :
-
1.0.0: Version initiale. -
1.0.1(par défaut) : champ de nom de tâche ajouté et ID fixe utilisé pour le brassage afin d'utiliser des ID stables.
-
Taille du téléchargement :
3.08 GiBTaille du jeu de données :
4.73 GiBMise en cache automatique ( documentation ): Non
Fractionnements :
| Diviser | Exemples |
|---|---|
'train' | 5 040 134 |
- Structure des fonctionnalités :
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),
})
- Documentation des fonctionnalités :
| Caractéristique | Classe | Façonner | Dtype | Description |
|---|---|---|---|---|
| FonctionnalitésDict | ||||
| définition | Texte | chaîne de caractères | ||
| identifiant | Texte | chaîne de caractères | ||
| contribution | Texte | chaîne de caractères | ||
| production | Texte | chaîne de caractères | ||
| la source | Texte | chaîne de caractères | ||
| Nom de la tâche | Texte | chaîne de caractères |
Clés supervisées (Voir
as_superviseddoc ):NoneFigure ( tfds.show_examples ) : non pris en charge.
Exemples ( tfds.as_dataframe ):
- Citation :
@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}
}