MSR2022 Encoding Study Data

该数据集是从一次实证研究中收集的，该研究比较了在多个系统和模型中用于软件性能学习的编码方案的性能。数据集涵盖了105个不同模型与编码方案组合的案例结果。任务是对用于软件性能学习的编码方案进行评估。

This dataset is collected from an empirical study whose core task is to evaluate encoding schemes for software performance learning. The study compares the performance of these encoding schemes across multiple systems and models, and the dataset encompasses the experimental results of 105 distinct combinations of models and encoding schemes.