A Python toolkit to download, load, and visualize predictive maintenance datasets.
Standardized Python toolkit for industrial PHM research and data loading
This toolkit provides a shared API and dataset-specific loaders for predictive maintenance research. It suits industrial data scientists and researchers working with anomaly detection and degradation data.
The repository lets you programmatically download, load, and visualize industrial datasets like CMAPSS using a unified Python API. It returns Polars DataFrames or LazyFrames depending on the dataset and includes built-in sensor trajectory visualization.
It suits machine learning engineers and researchers benchmarking prognostics and health management algorithms. It does not suit practitioners looking for a turnkey production anomaly detection service.
Access multiple predictive maintenance datasets through a consistent loading interface.
Data loaders return high-performance Polars DataFrames or LazyFrames.
Generate sensor trajectory figures directly from loaded data frames.
Importing or loading never triggers downloads, allowing you to fetch only what you need.
The repository code and documentation are free under the MIT License, though individual datasets remain subject to their own terms and source licenses. The only cost is running it yourself.
880 open remote AI-training roles, updated November 15, 2026; most pay $20–$175 an hour. Zekai lists them free; the application goes to the employer’s platform.
These reviewed AI tools do the same job instantly — from our scored directory.
No, importing or loading never downloads data; you must call the download function explicitly.
Loaders return Polars DataFrames or LazyFrames depending on the specific dataset.
Consult the usage guide in the docs directory for path configuration and Kaggle auth setup.
See Zekai first in Google