withmartian/cs13_dataset_100k_processed

这是一个包含表格数据、SQL指令和判断结果的数据集，适用于训练和评估自然语言处理模型在SQL生成和验证方面的性能。数据集包含训练集、验证集和测试集，每个集合都有相应的字节数和示例数。数据集字段包括表格名称、指令短语、列名、选择的列、列数据类型、SQL提示、SQL上下文、SQL语句、SQL有效性、SQL有效性验证消息、LLM判断、文本到SQL的LLM判断结果等。

This dataset contains table data, SQL instructions, and judgment results, which is suitable for training and evaluating the performance of natural language processing models in SQL generation and validation. The dataset includes training, validation, and test sets, each with corresponding byte sizes and example counts. Dataset fields include table names, instruction phrases, column names, selected columns, column data types, SQL prompts, SQL contexts, SQL statements, SQL validity, SQL validity validation messages, LLM judgments, and text-to-SQL LLM judgment results.