cwolff/SQaLe_2_Split
收藏Hugging Face2026-04-26 更新2026-05-03 收录
下载链接：
https://hf-mirror.com/datasets/cwolff/SQaLe_2_Split
下载链接
链接失效反馈官方服务：
资源简介：
该数据集是一个用于自然语言到SQL转换任务的大规模数据集，包含问题、对应的SQL语句、难度级别、问题风格、问题ID、模式ID、完整数据库模式、模式内容、相关表、相关表数量以及执行结果等特征。数据集分为训练集和测试集，训练集有1,753,924个示例，测试集有92,848个示例，适用于训练和评估SQL生成模型。
This dataset is a large-scale dataset for natural language to SQL conversion tasks, containing features such as questions, corresponding SQL statements, difficulty levels, question styles, question IDs, schema IDs, full database schemas, schema content, relevant tables, number of relevant tables, and execution results. The dataset is split into training and test sets, with 1,753,924 examples in the training set and 92,848 examples in the test set, suitable for training and evaluating SQL generation models.
提供机构：
cwolff
