qasc

  • Descrizione :

QASC è un set di dati con risposta a domande incentrato sulla composizione delle frasi. Consiste in 9.980 domande a scelta multipla a 8 vie sulla scienza della scuola elementare (8.134 treno, 926 dev, 920 test) e viene fornito con un corpus di 17 milioni di frasi.

Diviso Esempi
'test' 920
'train' 8.134
'validation' 926
  • Struttura delle caratteristiche :
FeaturesDict({
    'answerKey': Text(shape=(), dtype=string),
    'choices': Sequence({
        'label': Text(shape=(), dtype=string),
        'text': Text(shape=(), dtype=string),
    }),
    'combinedfact': Text(shape=(), dtype=string),
    'fact1': Text(shape=(), dtype=string),
    'fact2': Text(shape=(), dtype=string),
    'formatted_question': Text(shape=(), dtype=string),
    'id': Text(shape=(), dtype=string),
    'question': Text(shape=(), dtype=string),
})
  • Documentazione delle funzionalità :
Caratteristica Classe Forma Tipo D Descrizione
CaratteristicheDict
tasto di risposta Testo corda
scelte Sequenza
scelte/etichetta Testo corda
scelte/testo Testo corda
combinatofatto Testo corda
fatto1 Testo corda
fatto2 Testo corda
domanda_formattata Testo corda
id Testo corda
domanda Testo corda
  • Citazione :
@article{allenai:qasc,
      author    = {Tushar Khot and Peter Clark and Michal Guerquin and Peter Jansen and Ashish Sabharwal},
      title     = {QASC: A Dataset for Question Answering via Sentence Composition},
      journal   = {arXiv:1910.11473v2},
      year      = {2020},
}