SParC

该数据集是一个跨领域的资料集，包含了大约4,300个问题序列，这些序列由超过12,000个问题-SQL配对组成。每个问题序列都源自Spider中的原始问题。该数据集的规模为4,300个序列，超过12,000个配对，其任务是对文本进行SQL转换。

This is a cross-domain dataset containing approximately 4,300 question sequences, which consist of over 12,000 question-SQL pairs. Each question sequence is derived from the original questions in the Spider dataset. The dataset comprises 4,300 sequences and more than 12,000 pairs, with its target task being text-to-SQL translation.