Question 8
Domain 1: Databricks Machine LearningIf a model tag was added by mistake, what is the cleanest fix?
Correct answer: A
Explanation
If a model tag was added by mistake, the cleanest fix is to remove only the tag and leave the model intact. This follows the principle of making the smallest necessary change: "Remove the tag without deleting the model."
Why each option is right or wrong
A. Remove the tag without deleting the model
The least disruptive correction is to undo only the erroneous tag assignment, because the model itself is not the problem and should remain intact. Deleting the model would be an unnecessary destructive action when the issue can be fixed by removing the tag alone, preserving all other data and relationships.
B. Delete the entire model version
Deleting a model version removes the resource itself, far beyond correcting an accidental label.
C. Retrain the model from scratch
Retraining changes model weights and outputs; a mistaken tag is only metadata, not a training issue.
D. Create a new workspace
A workspace is a broader environment boundary and does not directly fix one incorrect model tag.