On the Relation between Code Quality and Machine Learning Performance: A Large-scale Empirical Study

The study examined 265,363 Kaggle notebooks to explore how code quality relates to machine learning performance. Using Pylint and SonarQube, it found that general Python code quality shows negligible correlation with performance, while ML‑specific violations have a small negative association with performance. Popularity and author expertise do not predict code quality or performance, though competition expertise correlates with better performance and fewer ML‑specific violations.

Machine-generated by The Flow from the publisher's headline and feed description — not written or checked by a human. The full article lives at arXiv Machine Learning.