Execution Time Dataset

该数据集包含了从四个实际Java开源程序中提取的922个代码片段，这些代码片段与标准化执行时间相关联。此外，该数据集源自开源Java项目，代表了实际测试执行时间的标准化数据，其时间范围被归一化至[0; 1]区间。数据涵盖了包括数据库、网络服务器和数据科学生命周期在内的多个应用领域。该数据集规模为922个测试文件，其任务是用图神经网络预测单元测试的执行时间。

This dataset contains 922 code snippets extracted from four real-world Java open-source projects, with each snippet paired with its standardized execution time. Derived from open-source Java projects, this dataset represents standardized data of actual test execution times, where all time values are normalized to the range [0, 1]. The data spans multiple application domains including databases, web servers, and data science lifecycles. This dataset consists of 922 test files, and the target task is to predict the execution time of unit tests using graph neural networks.