CodeGenEvaluation

CodeScore: Evaluating Code Generation by Learning Code Execution (Accepted to TOSEM)

Paper, Code, Model, Dataset

Extended test cases

Huggingface Datasets:

• HumanEval-ET
• MBPP-ET

Google Drive:

• APPS-ET

Citation

@article{dong2023codescore,
  title={Codescore: Evaluating code generation by learning code execution},
  author={Dong, Yihong and Ding, Jiazheng and Jiang, Xue and Li, Ge and Li, Zhuo and Jin, Zhi},
  journal={arXiv preprint arXiv:2301.09043},
  year={2023}
}