Annoy-PyEdu-Rs-Raw
收藏资源简介:
该数据集为PythonEdu-Rs的原始处理数据,源自HuggingFaceTB团队的数据集。数据以JSONL格式存储,每一行代表一个样本,包含以下字段:问题描述(problem_description)、输入输出要求与约束(io_requirements)、参考代码(refcode,包括导入包、辅助函数和主入口函数)、函数名(funcname)、输入输出示例列表(ios,每个元素包含输入参数input和返回值output)、原始代码来源(source)、推理类别(category)以及元信息(meta)。部分样本的ios字段为空,这是因为执行时输入输出规模过大,超出约束条件而未存储。注意:由于基于大语言模型的转换不够完善,部分问题描述未能充分描述代码,这将在未来版本中得到改进。该数据集适用于代码生成、程序合成、函数理解等任务,遵循Open Data Commons Attribution License (ODC-By v1.0)许可。
This dataset is the raw processed data of PythonEdu-Rs, originally from the HuggingFaceTB teams dataset. The data is stored in JSONL format, with each line representing a sample containing the following fields: problem_description, io_requirements, refcode (including import packages, helper functions, and main entry function), funcname, ios (a list of input-output examples, each element containing input parameters and return value), source, category, and meta. Some samples have an empty ios field because the input-output scale during execution was too large and exceeded the constraint conditions, so it was not stored. Note: Due to the imperfect conversion based on large language models, some problem descriptions fail to fully describe the code, which will be improved in future versions. This dataset is suitable for tasks such as code generation, program synthesis, and function understanding, and is licensed under the Open Data Commons Attribution License (ODC-By v1.0).




