- توضیحات :
مجموعه داده منطق فیلم حاوی دلایل منطقی مشروح شده توسط انسان برای بررسی فیلم است.
صفحه اصلی : http://www.cs.jhu.edu/~ozaidan/rationales/
کد منبع :
tfds.text.MovieRationalesنسخه ها :
-
0.1.0(پیش فرض): بدون یادداشت انتشار.
-
حجم دانلود :
3.72 MiBحجم مجموعه داده :
8.37 MiBذخیره خودکار ( اسناد ): بله
تقسیم ها :
| شکاف | مثال ها |
|---|---|
'test' | 199 |
'train' | 1600 |
'validation' | 200 |
- ساختار ویژگی :
FeaturesDict({
'evidences': Sequence(Text(shape=(), dtype=string)),
'label': ClassLabel(shape=(), dtype=int64, num_classes=2),
'review': Text(shape=(), dtype=string),
})
- مستندات ویژگی :
| ویژگی | کلاس | شکل | نوع D | شرح |
|---|---|---|---|---|
| FeaturesDict | ||||
| شواهد | دنباله (متن) | (هیچ یک،) | رشته | |
| برچسب | ClassLabel | int64 | ||
| مرور | متن | رشته |
کلیدهای نظارت شده (به
as_superviseddoc مراجعه کنید):Noneشکل ( tfds.show_examples ): پشتیبانی نمی شود.
مثالها ( tfds.as_dataframe ):
- نقل قول :
@unpublished{eraser2019,
title = {ERASER: A Benchmark to Evaluate Rationalized NLP Models},
author = {Jay DeYoung and Sarthak Jain and Nazneen Fatema Rajani and Eric Lehman and Caiming Xiong and Richard Socher and Byron C. Wallace}
}
@InProceedings{zaidan-eisner-piatko-2008:nips,
author = {Omar F. Zaidan and Jason Eisner and Christine Piatko},
title = {Machine Learning with Annotator Rationales to Reduce Annotation Cost},
booktitle = {Proceedings of the NIPS*2008 Workshop on Cost Sensitive Learning},
month = {December},
year = {2008}
}