This is a dedicated watch page for a single video.
You're setting up a component-based training pipeline in Azure Machine Learning. The pipeline includes several custom components that must pass parameters and configurations to each other. What approach should you adopt to define these parameters effectively for each job run within the pipeline?