Bearing diagnosis: accuracy, robustness and transfer — reproducibility software release
Reproducibility software release for the study ‘Separating diagnostic accuracy from robustness and transfer in bearing fault diagnosis’. The archive contains the preprocessing, feature extraction, training, evaluation, partition, prediction-replay and verification code used for the manuscript revision. It is intended to support reproducibility of the reported public-data experiments. Repository: https://github.com/YYB779/bearing-diagnosis-accuracy-robustness-transferRelease: https://github.com/YYB779/bearing-diagnosis-accuracy-robustness-transfer/releases/tag/v2026.11.21 First-party software is released under the MIT License. Third-party datasets, derived materials, model weights and external dependencies retain their own terms; the archive does not redistribute restricted raw datasets.
Authors
- Zongyuan Yang (ORCID: https://orcid.org/0009-0003-3798-3565)
Publication Details
- Journal
- Zenodo (CERN European Organization for Nuclear Research)
- Published
- 2026-11-21
- DOI
- https://doi.org/10.5281/zenodo.24427045
- Primary Topic
- Machine Fault Diagnosis Techniques
- Type
- article
- Field-Weighted Citation Impact
- 0.00