stefanocarrera/autophagycode_D_he_train-mercury_Qwen3-8B_strategy_trust_t1.25_g4_run2_metrics
收藏资源简介:
该数据集是一个用于代码分析和评估的数据集,包含164个训练样本,每个样本具有多个特征字段,如任务ID(task_id)、入口点(entry_point)、是否可执行(is_executable)、是否正确(is_correct)、测试通过数(tests_passed)、测试失败数(tests_failed)、测试运行时间(test_run_time_ms)、错误类型(error_type)、Halstead复杂度指标(如词汇量、长度、体积、难度、努力程度、时间)、圈复杂度(cyclomatic_complexity)、可维护性指数(maintainability_index)、代码行数(loc、sloc)、注释百分比(comment_percentage)、类型标记比(TTR)、令牌字典(token_dict)、香农熵(shannon_entropy)、平均预测熵(mean_predictive_entropy)、最大预测熵(max_predictive_entropy)、定义函数数(n_func_defined)和入口点重复性(entry_point_repeated)。这些特征用于评估代码质量、测试执行结果和软件工程度量,适用于代码测试、质量保证和研究目的。
This dataset is designed for code analysis and evaluation, consisting of 164 training examples. Each example includes multiple features such as task_id, entry_point, is_executable, is_correct, tests_passed, tests_failed, test_run_time_ms, error_type, Halstead complexity metrics (e.g., vocabulary, length, volume, difficulty, effort, time), cyclomatic_complexity, maintainability_index, lines of code (loc, sloc), comment_percentage, type-token ratio (TTR), token_dict, shannon_entropy, mean_predictive_entropy, max_predictive_entropy, number of functions defined (n_func_defined), and entry_point_repeated. These features are used to assess code quality, test execution outcomes, and software engineering metrics, making it suitable for code testing, quality assurance, and research purposes.




