Welcome to DSML4s8e’s documentation!
The library name is a abbreviation of Data Science / ML flow standalone.
Dsml4s8e is a Python library that extends Dagster to help manage the ML workflow around Jupyter notebooks development.
- DSML4s8e addresses issues:
Building of pipelines from standalone notebooks
Standardizing a structure of ML/DS pipeline projects to easy share and continuous improve them
Managing pipelines data in a continuous improvement process of its
- Dsml4s8e designed to support the following workflow:
Define a project structure and a structure of the pipeline data catalog for your pipeline by using class Storage Catalog ABC.
Develop standalone Jupyter notebooks corresponding specification requirements.
Difine a pipeline – a sequence of notebooks and deloy the pipelien in vary environments(experimental/test/prod).
Execute pipelines many times with different configurations in vary environments and on vary infrastructure..
Check out the Usage section for further information, including how to Installation the project.
Note
This project is under active development.