autophagycode_metrics_D_metrics_he_unsloth__Qwen3-14B-Base-unsloth-bnb-4bit_lr0.0001_gen10

该数据集包含164个训练样本，总大小为17,961字节，记录了编程任务的执行特征与代码质量指标。每个样本包含14个结构化字段：1) 任务标识（task_index）和入口点（entry_point）；2) 执行状态（is_executable, is_correct）及测试结果（tests_passed, tests_failed, test_run_time_ms）；3) 错误类型（error_type）；4) 代码复杂度指标（halstead_vocabulary, halstead_length, halstead_volume, halstead_difficulty, halstead_effort）和可维护性指数（maintainability_index）。数据集适用于代码质量分析、程序错误检测和软件度量研究等领域。

This dataset contains 164 training samples with a total size of 17,961 bytes, which records the execution features of programming tasks and code quality metrics. Each sample includes 14 structured fields: 1) Task identifier (task_index) and entry point (entry_point); 2) Execution status (is_executable, is_correct) and test results (tests_passed, tests_failed, test_run_time_ms); 3) Error type (error_type); 4) Code complexity metrics (halstead_vocabulary, halstead_length, halstead_volume, halstead_difficulty, halstead_effort) and maintainability index (maintainability_index). This dataset is applicable to research fields such as code quality analysis, program error detection and software metric studies.