MLPER-02: Use purpose-built AI and ML services and resources
Consider how part or all of the workload could be handled by pre-built AI services or ML resources. Better performance can often be delivered more efficiently by using pre-optimized components included in AI and ML managed services. Select an optimal mix of bespoke and pre-built components to meet the workload requirements.
Implementation plan
-
Learn about AWS AI services - Determine whether AWS managed AI services
are applicable to the business use case. Understand how managed AWS AI services can relieve the burden of training and maintaining an ML pipeline. Use Amazon SageMaker AI to develop in the cloud and understand the roles and responsibilities needed to maintain the ML workload. Consider combining managed AI services with custom ML models built on Amazon SageMaker AI. This approach allows balancing the tradeoffs between ML workload management, and solutions specificity for the business use case. -
Learn about SageMaker AI JumpStart – This service provides pre-trained, open-source models for a wide range of problem types to help you get started with machine learning.
-
Learn about SageMaker AI Algorithms and Models in AWS Marketplace, a curated digital catalog that makes it easy for you to find, buy, deploy, and manage third-party software and services that can help you build solutions and run their businesses.
Documents
Videos
