We build upon the introductory DVC demo and add the tracking of ML artefact lineage from data and the underlying ML pipeline source code.
- How to initialise a DVC pipelines project.
- How to develop a ML pipeline for use with DVC.
- How to run a ML pipeline via DVC and track the resulting artefacts.
- How to retrieve pipeline run metrics.
Running the Demo¶
This demo is contained within a single Jupyter notebook -
demos/dvc-pipelines/dvc_pipelines.ipynb. Make sure you have the necessary Python package requirements installed into a Jupyter kernel for it to run successfully.