ML Metadata

Conceptual overview about Metadata in Kubeflow Pipelines

Note: Kubeflow Pipelines has moved from using kubeflow/metadata to using google/ml-metadata for Metadata dependency.

Kubeflow Pipelines backend stores runtime information of a pipeline run in Metadata store. Runtime information includes the status of a task, availability of artifacts, custom properties associated with Execution or Artifact, etc. Learn more at ML Metadata Get Started.

You can view the connection between Artifacts and Executions across Pipeline Runs, if one Artifact is being used by multiple Executions in different Runs. This connection visualization is called a Lineage Graph.

Next steps

Feedback

Was this page helpful?


Last modified May 31, 2024: Moved "Concepts" to Pipelines (9ea2e537)