Answer-first summary for fast verification
Answer: No
The proposed solution to add the `scoring_script` parameter does NOT meet the goal. For MLflow models with PyFunc flavor deployed to online endpoints in Azure Machine Learning, no-code deployment is supported, meaning neither a scoring script nor an environment needs to be specified. This is explicitly stated in the Microsoft documentation and confirmed by community discussions with upvotes. The requirement for no egress connectivity is addressed through model packaging (using the `--with-package` flag in CLI or equivalent in SDK), not by adding a scoring script. Adding a scoring script is unnecessary for MLflow models and does not resolve the egress connectivity constraint.
Author: LeetQuiz Editorial Team
Ultimate access to all questions.
No comments yet.
You have an Azure Machine Learning workspace named Workspace1 containing a registered MLflow model named model1 with the PyFunc flavor. You intend to deploy model1 to an online endpoint named endpoint1 that has no egress connectivity, using the Azure Machine Learning Python SDK v2.
You are provided with the following code:
# Code block for deployment
# Code block for deployment
You must add a parameter to the ManagedOnlineDeployment object to ensure the model deployment is successful.
Proposed Solution: Add the scoring_script parameter.
Does the proposed solution meet the goal?
A
Yes
B
No