PELAB-LiU/JunoBench
收藏Hugging Face2025-10-22 更新2025-11-01 收录
下载链接：
https://hf-mirror.com/datasets/PELAB-LiU/JunoBench
下载链接
链接失效反馈官方服务：
资源简介：
JunoBench是一个针对机器学习Python Jupyter笔记本中崩溃的基准数据集。该数据集包含111个可重现的带有崩溃的ML笔记本，每个笔记本都配有一个可验证的修复版本。它涵盖了多种流行的机器学习库，包括TensorFlow/Keras、PyTorch、Scikit-learn、Pandas和NumPy，以及特定于笔记本的执行顺序错误。JunoBench提供了一个统一的执行环境，可以可靠地重现所有崩溃和修复。
JunoBench is a benchmark dataset of crashes in Machine Learning Python Jupyter Notebooks. This dataset contains 111 reproducible ML notebooks with crashes, each paired with a verifiable fix. It covers a wide range of popular ML libraries, including TensorFlow/Keras, PyTorch, Scikit-learn, Pandas, and NumPy, as well as notebook-specific out-of-order execution issues. JunoBench provides a unified execution environment in which all crashes and fixes can be reliably reproduced.
提供机构：
PELAB-LiU
