Question 80
A pipeline graph has a preprocessing stage and training stage but the training input points directly to raw data. The next test must distinguish a real fix from a configuration change that leaves the symptom intact. The team needs evidence that the training run uses processed rows from the upstream component. Which implementation is the best fit for this requirement?
Explanation
Correct Answer
Explicit graph edges determine which staged artifact is consumed by a downstream step. After the change, the training run uses processed rows from the upstream component.
Incorrect Answers
Component interface types distinguish data references from scalar parameters. Explicit graph edges determine which staged artifact is consumed by a downstream step.
A component defines code, command, environment, and interface for controlled pipeline reuse.
Component input-output contracts determine how data assets are routed across pipeline nodes.