AWS Certified Data Analytics Specialty 2021 – Hands-On - SageMaker

Amazon SageMaker is a powerful, fully managed machine learning service on AWS. It offers modules for building, training, and deploying models, supporting various frameworks like TensorFlow and PyTorch. SageMaker provides scalable algorithms, reinforcement learning, and hyperparameter tuning. It ensures security with encryption and integrates with AWS services like CloudWatch. While it requires Python proficiency, it enables advanced AI systems with cloud and edge deployment capabilities.

What is a key advantage of Amazon SageMaker over Amazon's older ML service?

• It does not require cloud resources.

• It is cheaper to use.

• It offers more flexibility and advanced algorithms.

• It requires no coding knowledge.

Which tool is used in SageMaker's build module for writing Python code?

• Apache Spark

• Jupyter Notebook

• RStudio

• Google Colab

What is the purpose of CUDA and CU DNN drivers in SageMaker?

• To provide data storage solutions

• To improve network security

• To accelerate neural networks using GPUs

• To enhance CPU performance

What does the training module in SageMaker allow you to do?

• Encrypt data at rest

• Monitor API calls

• Perform one-click model training and tuning

• Create data visualizations

What is SageMaker Neo used for?

• To monitor SageMaker usage

• To encrypt data in transit

• To run models in the cloud and at the edge

• To train models on local machines

How does SageMaker ensure the security of your code?

• By allowing public access

• By storing it on local machines

• By using security groups and optional encryption

• By storing it in unencrypted volumes

Which service is used to monitor Amazon SageMaker?

• Amazon RDS

• Amazon EC2

• Amazon CloudWatch

• Amazon S3