The new service, Amazon SageMaker Pipelines, has been launched to provide continuous integration and delivery pipelines that automate steps of ML (Machine Learning) workflows. It’s the first CI/CD service for ML to build, store, and track automated workflows and also create an audit trail for training data and modeling configurations.
Why is it important?
- Ease of Use: It has built in ML workflow templates that can be used to build, test, and deploy ML models quickly.
- Compliance: Amazon SageMaker pipeline logs can be saved as audit trails to recreate models for similar future business cases that help support compliance requirements.
- Better Capabilities: This service brings CI/CD practices to ML, which keeps the development & production environments separate, version control measures, on-demand testing, and end-to-end automation.
- Automation: As the first Purpose-built CI/CD service for ML that incorporates automation of data loading, transformation, training, tuning, and deployment workflow steps, this increases productivity significantly.
- Scalability: With the ability to create, automate, and manage end-to-end ML workflows at scale, there’s peace of mind knowing various are stored and be referred back to for audit purposes, compliance requirements, and future solution builds.
Why We’re Excited
Amazon SageMaker Pipelines offer a more efficient and productive solution to scale by reusing the workflow steps created and stored in a central repository. With built-in templates to deploy, train, and test models, our ML teams can quickly leverage CI/CD in our ML environments and easily incorporate models we’ve already created. With the SageMaker Pipelines model registry, we can track these model versions in one central location that gives us visibility and up to date record logs of the best possible solution options to meet client deployment needs.
If you’re looking to explore these services further and need some guidance, let us know and we’ll connect you to an Idexcel expert!