Answer-first summary for fast verification
Answer: Yes
The solution meets the goal because the `with_package` parameter is specifically designed for deploying MLflow models to online endpoints without egress connectivity. This parameter enables model packaging, which bundles all necessary dependencies with the model, eliminating the need for internet connectivity during deployment to download Python packages. The community discussion strongly supports this with multiple comments (including highly upvoted ones) confirming that `with_package=True` is the correct approach for this scenario. The alternative answer (B) suggesting that scoring_script is required is incorrect for MLflow PyFunc models, as MLflow models contain their own scoring logic and don't require external scoring scripts.
Author: LeetQuiz Editorial Team
Ultimate access to all questions.
You have an Azure Machine Learning workspace named Workspace1 that contains a registered MLflow model named model1 with the PyFunc flavor. You plan to deploy model1 to an online endpoint named endpoint1 without egress connectivity using the Azure Machine Learning Python SDK v2.
You have the following code:
# Code for creating a ManagedOnlineDeployment
# Code for creating a ManagedOnlineDeployment
You need to add a parameter to the ManagedOnlineDeployment object to ensure the model deploys successfully.
Proposed Solution: Add the with_package parameter.
Does the solution meet the goal?
A
Yes
B
No
No comments yet.