stefanocarrera/autophagycode_D_he_train-mercury_Qwen3-8B_strategy_trust_t1.25_g7_run0_metrics
收藏资源简介:
该数据集包含164个样本,每个样本包含代码任务ID、入口点、是否可执行、是否正确、测试通过/失败数量、测试运行时间、错误类型,以及多种代码度量指标,包括Halstead复杂度(词汇量、长度、体积、难度、工作量、时间)、圈复杂度、可维护性指数、代码行数、源代码行数、注释百分比、词型-词例比(TTR)、词例字典、香农熵、平均预测熵、最大预测熵、定义函数数量、入口点是否重复。这些数据可用于分析代码执行结果与代码质量之间的关系。
This dataset contains 164 samples, each including task ID, entry point, whether the code is executable, whether it is correct, number of tests passed/failed, test run time, error type, and various code metrics: Halstead measures (vocabulary, length, volume, difficulty, effort, time), cyclomatic complexity, maintainability index, lines of code, source lines of code, comment percentage, type-token ratio (TTR), token dictionary, Shannon entropy, mean predictive entropy, max predictive entropy, number of functions defined, and whether the entry point is repeated. These data are useful for analyzing the relationship between code execution outcomes and code quality.




