MLA-C01 Question 13
Select 4You are building a machine learning model on AWS to predict customer churn for a subscription-based service. You decide to use Amazon SageMaker for the end-to-end workflow. Which of the following steps can be performed directly within SageMaker Studio to facilitate your machine learning pipeline?
- A
Data preprocessing and feature engineering using built-in Jupyter notebooks
- B
Training and tuning the model using SageMaker's built-in algorithms
- C
Scheduling automated deployments using Amazon EventBridge
- D
Evaluating the model’s performance with built-in metrics visualization
- E
Monitoring deployed ML models for data drift using SageMaker Model Monitor
Show answer and explanation
Correct answers: A, B, D, E
Explanation
Amazon SageMaker Studio is a fully integrated development environment that allows machine learning engineers to perform tasks such as data preprocessing, model training, evaluation, and monitoring in a unified interface. While SageMaker Studio provides extensive end-to-end capabilities for ML pipelines, certain tasks like scheduling deployments require the use of additional AWS services such as EventBridge.
- A. Correct.
Correct. SageMaker Studio provides Jupyter notebooks that allow you to perform data preprocessing and feature engineering directly within the environment.
- B. Correct.
Correct. SageMaker supports training and hyperparameter tuning with built-in algorithms, which you can manage directly within SageMaker Studio.
- C. Incorrect.
Incorrect. Scheduling automated deployments is not directly supported within SageMaker Studio. Amazon EventBridge is a separate service used for scheduling and orchestrating tasks.
- D. Correct.
Correct. SageMaker Studio provides tools for evaluating and visualizing model performance metrics during and after training.
- E. Correct.
Correct. SageMaker Model Monitor can be used to monitor deployed models for issues like data drift, and it integrates seamlessly with SageMaker Studio.