stefanocarrera/autophagycode_D_he_train-mercury_Qwen3-8B_strategy_trust_t1.25_g4_run1_metrics
收藏资源简介:
该数据集包含代码任务的评估指标,涉及任务ID、执行点、可执行状态、正确性、测试通过与失败数量、错误类型,以及多种代码复杂度度量(如Halstead词汇量、长度、体积、难度、努力程度和时间,圈复杂度,可维护性指数),代码规模指标(如代码行数、有效代码行数、注释比例),词汇多样性(TTR)、熵值(香农熵、预测熵均值和最大值),以及函数定义数量和入口点重复性。数据集适用于代码质量分析、自动化测试验证或程序评估研究。
This dataset includes evaluation metrics for code tasks, covering task ID, entry point, executable status, correctness, number of tests passed and failed, error type, and various code complexity measures (e.g., Halstead vocabulary, length, volume, difficulty, effort, and time, cyclomatic complexity, maintainability index), code size metrics (e.g., lines of code, source lines of code, comment percentage), lexical diversity (TTR), entropy values (Shannon entropy, mean and max predictive entropy), number of function definitions, and entry point repetition. The dataset is suitable for code quality analysis, automated test validation, or program evaluation research.




