ml-pipeline-importer-runner

Creates or uses an existing OpenShift AI data science pipeline, and run's it.

example for pipeline creation can be seen in this repo: populate-vectors-pipeline

Mandatory Env Variables

PIPELINE_YAML_RAW_URL

Defines the RAW URL of the compiled KFP pipeline program.

Optional Env Variables (with Defaults)

PIPELINE_NAME

Specifies the name of the pipeline.

Default Name: Default

EXPERIMENT_NAME

Specifies the name of the experiment.

Default Name: Default

RUN_NAME

Specifies the name of the run.

Default: 1

PIPELINE_PARAMS

Additional parameters for the pipeline in JSON dict string format.

Example: '{"load_from_repo": True, "load_from_s3": False, "load_from_urls": False}'