snli ، snli

  • الوصف :

مجموعة SNLI (الإصدار 1.0) عبارة عن مجموعة من 570 ألف زوج من الجمل الإنجليزية المكتوبة من قبل الإنسان والتي تم تصنيفها يدويًا من أجل التصنيف المتوازن مع التسميات التي تستلزم ، والتناقض ، والحيادية ، مما يدعم مهمة الاستدلال اللغوي الطبيعي (NLI) ، والمعروف أيضًا بالتعرف على الاستدلال النصي (RTE).

انشق، مزق أمثلة
'test' 10000
'train' 550152
'validation' 10000
  • هيكل الميزة :
FeaturesDict({
    'hypothesis': Text(shape=(), dtype=string),
    'label': ClassLabel(shape=(), dtype=int64, num_classes=3),
    'premise': Text(shape=(), dtype=string),
})
  • وثائق الميزة :
ميزة فصل شكل نوع وصف
الميزات
فرضية نص سلسلة
ضع الكلمة المناسبة ClassLabel int64
فرضية نص سلسلة
  • الاقتباس :
@inproceedings{snli:emnlp2015,
    Author = {Bowman, Samuel R. and Angeli, Gabor and Potts, Christopher, and Manning, Christopher D.},
    Booktitle = {Proceedings of the 2015 Conference on Empirical Methods in Natural Language Processing (EMNLP)},
    Publisher = {Association for Computational Linguistics},
    Title = {A large annotated corpus for learning natural language inference},
    Year = {2015}
}