nerdexam
Amazon

MLS-C01 · Question #360

An agriculture company wants to improve crop yield forecasting for the upcoming season by using crop yields from the last three seasons. The company wants to compare the performance of its new scikit-

The correct answer is A. Package the code as the training script for an Amazon SageMaker scikit-learn container.. To compare a new scikit-learn model with a benchmark, run the code within a container, and offload operational maintenance to AWS, the best approach is to package the code as a script for an Amazon SageMaker scikit-learn container.

Machine Learning Implementation and Operations

Question

An agriculture company wants to improve crop yield forecasting for the upcoming season by using crop yields from the last three seasons. The company wants to compare the performance of its new scikit-learn model to the benchmark. A data scientist needs to package the code into a container that computes both the new model forecast and the benchmark. The data scientist wants AWS to be responsible for the operational maintenance of the container. Which solution will meet these requirements?

Options

  • APackage the code as the training script for an Amazon SageMaker scikit-learn container.
  • BPackage the code into a custom-built container. Push the container to Amazon Elastic Container
  • CPackage the code into a custom-built container. Push the container to AWS Fargate.
  • DPackage the code by extending an Amazon SageMaker scikit-learn container.

How the community answered

(22 responses)
  • A
    73% (16)
  • B
    9% (2)
  • C
    14% (3)
  • D
    5% (1)

Why each option

To compare a new scikit-learn model with a benchmark, run the code within a container, and offload operational maintenance to AWS, the best approach is to package the code as a script for an Amazon SageMaker scikit-learn container.

APackage the code as the training script for an Amazon SageMaker scikit-learn container.Correct

Packaging the comparison logic as a script and running it with an Amazon SageMaker scikit-learn container leverages AWS-managed infrastructure and dependencies, ensuring that AWS is responsible for the operational maintenance of the container while allowing custom code execution for both the new model and benchmark.

BPackage the code into a custom-built container. Push the container to Amazon Elastic Container

While Amazon ECR stores custom-built containers, deploying and managing these containers for execution would still require user-managed compute services (like EC2, ECS, or EKS), placing significant operational maintenance responsibility on the user rather than AWS.

CPackage the code into a custom-built container. Push the container to AWS Fargate.

Pushing a custom-built container to AWS Fargate offers serverless compute for container orchestration, but the operational maintenance of the ML environment within the container and its framework dependencies remains the user's responsibility, unlike a SageMaker managed container.

DPackage the code by extending an Amazon SageMaker scikit-learn container.

Extending an Amazon SageMaker scikit-learn container implies building a custom Docker image on top of a base image, which increases the user's responsibility for maintaining the custom layers and their dependencies, thus not fully offloading operational maintenance to AWS.

Concept tested: SageMaker pre-built containers and managed services

Source: https://docs.aws.amazon.com/sagemaker/latest/dg/sagemaker-scikit-learn.html

Topics

#Amazon SageMaker#Scikit-learn#Containerization#Managed ML Services

Community Discussion

No community discussion yet for this question.

Sign up to join the discussion
Full MLS-C01 Practice