zhenzhen123333/CSQA_preprocessed

CSQA_preprocessed是一个预处理后的常识问答数据集，用于多项选择问答任务。数据集包含id、问题（question）、问题概念（question_concept）、选项列表（choices，每个选项有标签和文本）、答案键（answerKey）、输入（inputs）和目标（targets）等特征。它分为训练集（9741个示例）、验证集（1221个示例）和测试集（1140个示例），旨在支持自然语言处理模型在常识推理方面的训练和评估。

CSQA_preprocessed is a preprocessed commonsense question answering dataset designed for multiple-choice question answering tasks. The dataset includes features such as id, question, question_concept, choices (each with label and text), answerKey, inputs, and targets. It is split into training (9741 examples), validation (1221 examples), and test (1140 examples) sets, aiming to support the training and evaluation of natural language processing models in commonsense reasoning.