openbookqa

  • Descrizione :

Il set di dati contiene 5.957 domande a scelta multipla a 4 vie. Inoltre, forniscono 5.167 fatti di conoscenza comune di crowdsourcing e una versione ampliata delle domande train/dev/test in cui ogni domanda è associata al suo fatto fondamentale di origine, un punteggio di accuratezza umana, un punteggio di chiarezza e un crowd-worker anonimo ID.

Diviso Esempi
'test' 500
'train' 4.957
'validation' 500
  • Struttura delle caratteristiche :
FeaturesDict({
    'answerKey': ClassLabel(shape=(), dtype=int64, num_classes=4),
    'clarity': float32,
    'fact1': Text(shape=(), dtype=string),
    'humanScore': float32,
    'question': FeaturesDict({
        'choice_A': Text(shape=(), dtype=string),
        'choice_B': Text(shape=(), dtype=string),
        'choice_C': Text(shape=(), dtype=string),
        'choice_D': Text(shape=(), dtype=string),
        'stem': Text(shape=(), dtype=string),
    }),
    'turkIdAnonymized': Text(shape=(), dtype=string),
})
  • Documentazione delle funzionalità :
Caratteristica Classe Forma Tipo D Descrizione
CaratteristicheDict
tasto di risposta ClassLabel int64
chiarezza Tensore galleggiante32
fatto1 Testo corda
humanScore Tensore galleggiante32
domanda CaratteristicheDict
domanda/scelta_A Testo corda
domanda/scelta_B Testo corda
domanda/scelta_C Testo corda
domanda/scelta_D Testo corda
domanda/radice Testo corda
turkIdAnonimizzato Testo corda
  • Citazione :
@article{mihaylov2018can,
  title={Can a suit of armor conduct electricity? a new dataset for open book question answering},
  author={Mihaylov, Todor and Clark, Peter and Khot, Tushar and Sabharwal, Ashish},
  journal={arXiv preprint arXiv:1809.02789},
  year={2018}
}