CodeXGLUE克隆检测-BigCloneBench数据集,用于训练模型以判断两个给定的Java方法是否为彼此的克隆。给定两个代码作为输入,任务是对其进行二元分类(0/1),其中1表示语义等价,0表示其他情况。模型通过F1分数进行评估。该数据集基于BigCloneBench,并按照论文《Detecting Code Clones with Graph Neural Network and Flo
The dataset of our work "A Comparative Study of Test Code Clones and Production Code Clones". It contains the dataset, the classified code clones, and all python scripts used during the postprocessing
This is the dataset of GPT-J code clone detection. 'data/' folder contains the data to replicate our study. 'code/' folder contains the code to replicate our study. `rebuttal/` folder cont
This is the dataset of GPT-J code clone detection. 'data/' folder contains the data to replicate our study. 'code/' folder contains the code to replicate our study. `rebuttal/` folder contains the res