qasc

  • Açıklama :

QASC, cümle kompozisyonuna odaklanan soru yanıtlayan bir veri kümesidir. İlkokul bilimiyle ilgili 9.980 8'li çoktan seçmeli sorudan (8.134 tren, 926 geliştirme, 920 test) oluşur ve 17 milyon cümlelik bir külliyatla birlikte gelir.

Bölmek örnekler
'test' 920
'train' 8.134
'validation' 926
  • Özellik yapısı :
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),
})
  • Özellik belgeleri :
Özellik Sınıf Şekil Dtipi Açıklama
ÖzelliklerDict
cevap anahtarı Metin sicim
seçimler Sıra
seçenekler/etiket Metin sicim
seçimler/metin Metin sicim
birleşik gerçek Metin sicim
gerçek1 Metin sicim
gerçek2 Metin sicim
biçimlendirilmiş_soru Metin sicim
İD Metin sicim
soru Metin sicim
  • Alıntı :
@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},
}