- תיאור:
גוף SNLI (גרסה 1.0) הוא אוסף של 570 אלף זוגות משפטים באנגלית בכתב-אנוש המסומנים באופן ידני לסיווג מאוזן עם התוויות הכרוכות, סתירות וניטרליות, התומכות במשימה של הסקת שפה טבעית (NLI), הידועה גם כזיהוי מעורבות טקסטואלית. (RTE).
קוד מקור:
tfds.text.Snli
גרסאות:
-
1.1.0
(ברירת המחדל): הערות השחרור לא.
-
גודל ההורדה:
90.17 MiB
מערך נתונים גודל:
87.00 MiB
Auto-במטמון ( תיעוד ): כן
פיצולים:
לְפַצֵל | דוגמאות |
---|---|
'test' | 10,000 |
'train' | 550,152 |
'validation' | 10,000 |
- מאפיינים:
FeaturesDict({
'hypothesis': Text(shape=(), dtype=tf.string),
'label': ClassLabel(shape=(), dtype=tf.int64, num_classes=3),
'premise': Text(shape=(), dtype=tf.string),
})
מפתחות השגחה (ראה
as_supervised
doc ):None
איור ( tfds.show_examples ): לא נתמך.
דוגמאות ( tfds.as_dataframe ):
- ציטוט:
@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}
}