ai-300 Practice Question 2015
Exam: ai-300
Domain: Implement CI/CD for machine learning
Topic: Automate model training and deployment
Difficulty: medium
You are tasked with automating the model training process using GitHub Actions for Azure Machine Learning. You want to ensure that the model is validated before it is deployed to production. Which of the following steps should you include in your GitHub Actions workflow to achieve this?
Answer Options
A
Train the model using Azure ML SDK and then deploy it directly to production.
B
Use Azure ML SDK to train the model, validate it using a test dataset, and then deploy it to production if the validation passes.
C
Deploy the model to production and then validate it using a test dataset.
D
Use Azure ML SDK to train the model and then manually validate it outside of the workflow.
Correct Answer
B: Use Azure ML SDK to train the model, validate it using a test dataset, and then deploy it to production if the validation passes.
Explanation
The correct answer is B because it includes both the training and validation steps within the same workflow. Option A only trains the model without validation, which does not meet the requirement. Option C only validates the model without training, which is not sufficient. Option D does not mention any validation or training steps, making it incorrect.