Published May 2025
| Version v5
Dataset
Open
Dataset of paper "Why do Machine Learning Notebooks Crash? An Empirical Study on Public Python Jupyter Notebooks"
Authors/Creators
-
1.
Linköping University
Description
All the related data of our paper "Why Do Machine Learning Notebooks Crash? An Empirical Study on Public Python Jupyter Notebooks" includes:
- GitHub and Kaggle notebooks that contain error outputs.
- GitHub notebooks are from The Stack repository [1].
- Kaggle notebooks are public notebooks on Kaggle platform from year 2023, downloaded via KGTorrent [2] with the Meta Kaggle dataset [3].
- Please refer to the sources for the licenses of the notebooks: The Stack repository [1] and Kaggle terms.
- Identified programming language results of GitHub notebooks.
- Identified ML library results from Kaggle notebooks.
- Datasets of crashes from GitHub and Kaggle.
- Clustering results of crashes from all crashes, and from GitHub and Kaggle respectively.
- Sampled crashes and associated notebooks (organized by cluster id).
- Manual labeling results.
The related code repository can be found here.
Files
data_jupyter_nbs_empirical.zip
Additional details
Software
- Repository URL
- https://github.com/PELAB-LiU/jupyter_nbs_empirical
- Programming language
- Python , Jupyter Notebook
References
- D. Kocetkov, R. Li, L. B. Allal, J. Li, C. Mou, C. M. Ferrandis, Y. Jernite, M. Mitchell, S. Hughes, T. Wolf, D. Bahdanau, L. von Werra, and H. de Vries, "The Stack: 3 TB of permissively licensed source code," in arXiv:2211.15533, 2022.
- Quaranta, Luigi, Calefato, Fabio, & Lanubile, Filippo. (2021). collab-uniba/KGTorrent: First release (v. 1.0.0) of KGTorrent (v1.0.0). Zenodo. https://doi.org/10.5281/zenodo.4472990
- M. Risdal and T. Bozsolik, "Meta Kaggle," 2022, https://www.kaggle.com/ds/9.