amazingvince/sythetic-scored-and-annotated

该数据集包含多个字段，涉及代码的原始内容、各项评分（如整体评分、功能性评分、可读性评分等）、评论（如功能性评论、可读性评论等）、优点、改进建议等。数据集的结构包括训练集的分割信息，如字节数、示例数等。配置信息包括默认配置的数据文件路径。

This dataset is designed for evaluating and analyzing code quality, featuring original code text, various scores (functionality, readability, efficiency, maintainability, error handling), corresponding comments for each score, strengths of the code, areas for improvement, suggestions, full JSON, snippet ID, requested score and length, and programming language. The dataset is split into a training set with 1080 examples.

原始信息汇总

数据集概述

数据集信息

特征

• original_code: 类型为 string
• overall_score: 类型为 int64
• functionality_score: 类型为 int64
• readability_score: 类型为 int64
• efficiency_score: 类型为 int64
• maintainability_score: 类型为 int64
• error_handling_score: 类型为 int64
• functionality_comments: 类型为 string
• readability_comments: 类型为 string
• efficiency_comments: 类型为 string
• maintainability_comments: 类型为 string
• error_handling_comments: 类型为 string
• strengths: 类型为 sequence，内容为 string
• areas_for_improvement: 类型为 sequence，内容为 string
• suggestions: 类型为 sequence，内容为 string
• full_json: 类型为 string
• snippet_id: 类型为 int64
• requested_score: 类型为 int64
• requested_length: 类型为 int64
• requested_programing_language: 类型为 string

数据分割

• train: 包含 1080 个样本，总字节数为 4660449

数据集大小

• 下载大小: 1746988 字节
• 数据集大小: 4660449 字节

配置

• default: 包含 train 分割的数据文件路径为 data/train-*