tasksource/ecqa

---
dataset_info:
features:
- name: q_no
dtype: string
- name: q_concept
dtype: string
- name: q_text
dtype: string
- name: q_op1
dtype: string
- name: q_op2
dtype: string
- name: q_op3
dtype: string
- name: q_op4
dtype: string
- name: q_op5
dtype: string
- name: q_ans
dtype: string
- name: taskA_pos
dtype: string
- name: taskA_neg
dtype: string
- name: taskB
dtype: string
splits:
- name: train
num_bytes: 6458760
num_examples: 7598
- name: validation
num_bytes: 924269
num_examples: 1090
- name: test
num_bytes: 1846056
num_examples: 2194
download_size: 5678604
dataset_size: 9229085
license: cdla-sharing-1.0
task_categories:
- question-answering
language:
- en
---
# Dataset Card for "ecqa"

https://github.com/dair-iitd/ECQA-Dataset
```
@inproceedings{aggarwaletal2021ecqa,
title={{E}xplanations for {C}ommonsense{QA}: {N}ew {D}ataset and {M}odels},
author={Shourya Aggarwal and Divyanshu Mandowara and Vishwajeet Agrawal and Dinesh Khandelwal and Parag Singla and Dinesh Garg},
booktitle="Proceedings of the 59th Annual Meeting of the Association for Computational Linguistics and the 11th International Joint Conference on Natural Language Processing (Volume 1: Long Papers)}",
year = "2021",
address = "Online",
publisher = "Association for Computational Linguistics"
}
```

原始信息汇总

数据集概述

数据集名称

• ECQA

数据集特征

• q_no: 数据类型为字符串
• q_concept: 数据类型为字符串
• q_text: 数据类型为字符串
• q_op1: 数据类型为字符串
• q_op2: 数据类型为字符串
• q_op3: 数据类型为字符串
• q_op4: 数据类型为字符串
• q_op5: 数据类型为字符串
• q_ans: 数据类型为字符串
• taskA_pos: 数据类型为字符串
• taskA_neg: 数据类型为字符串
• taskB: 数据类型为字符串

数据集划分

• 训练集: 包含7598个示例，占用6458760字节
• 验证集: 包含1090个示例，占用924269字节
• 测试集: 包含2194个示例，占用1846056字节

数据集大小

• 下载大小: 5678604字节
• 数据集总大小: 9229085字节

许可证

• CDLA-Sharing-1.0

任务类别

• 问答

语言

• 英语