Created by: gsvigruha
What is this Python project?
A build system for data science pipelines. It can build model artifacts from version controlled data and code, keep track of dependencies and rerun the necessary parts of the pipeline upon changes.
What's the difference between this Python project and similar ones?
- You can version both code and data simultaneously.
- Intermediate data, model files, test results are all build targets.
- Build target version hashcodes are precomputed from the dependencies and all results are cached, so Prodmodel only executes what's strictly necessary.
--
Anyone who agrees with this pull request could vote for it by adding a