tasksource/ReSQ

数据集ReSQ包含训练集和开发集，每个数据集包含故事、问题、问题类型、答案和候选答案等特征。训练集包含2450个样本，开发集包含663个样本。数据集的下载大小为58830字节，总大小为499998字节。该数据集在2022年EMNLP会议上被用于空间角色标注和推理的研究。The dataset ReSQ consists of a training set and a development set, both of which include features such as story, question, question type, answer, and candidate answers. The training set contains 2450 samples, while the development set contains 663 samples. The download size of the dataset is 58,830 bytes, and the total size is 499,998 bytes. This dataset was employed for research on spatial role labeling and reasoning at the 2022 Conference on Empirical Methods in Natural Language Processing (EMNLP).

原始信息汇总

数据集概述

数据集配置

• 配置名称: default
• 数据文件:
  • 训练集: data/train-*
  • 验证集: data/dev-*

数据集信息

• 特征:
  • story: 数据类型为字符串
  • question: 数据类型为字符串
  • q_type: 数据类型为字符串
  • answer: 数据类型为字符串序列
  • candidate_answers: 数据类型为字符串序列
• 数据分割:
  • 训练集:
    • 字节数: 388340
    • 样本数: 2450
  • 验证集:
    • 字节数: 111658
    • 样本数: 663
• 下载大小: 58830
• 数据集大小: 499998