snli

The SNLI corpus (version 1.0) is a collection of 570k human-written English sentence pairs manually labeled for balanced classification with the labels entailment, contradiction, and neutral, supporting the task of natural language inference (NLI), also known as recognizing textual entailment (RTE).

URL: https://nlp.stanford.edu/projects/snli/
DatasetBuilder: tfds.text.snli.Snli

snli is configured with tfds.core.dataset_builder.BuilderConfig and has the following configurations predefined (defaults to the first one):

plain_text (v1.0.0) (Size: 90.17 MiB): Plain text import of SNLI

`snli/plain_text`

Plain text import of SNLI

Versions:

1.0.0 (default): New split API (https://tensorflow.org/datasets/splits)
0.0.1: None

Statistics

Split	Examples
ALL	570,152
TRAIN	550,152
TEST	10,000
VALIDATION	10,000

Features

FeaturesDict({
    'hypothesis': Text(shape=(), dtype=tf.string),
    'label': ClassLabel(shape=(), dtype=tf.int64, num_classes=3),
    'premise': Text(shape=(), dtype=tf.string),
})

Homepage

https://nlp.stanford.edu/projects/snli/

Citation

@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}
}