AWS Sagemaker
AWS Tutorial: Machine Learning on AWS (SageMaker)
Welcome to the Amazon SageMaker lesson. If your team consists of expert Data Scientists who want to build custom machine learning models from scratch, this is the tool they use.
Why Learn Amazon SageMaker?
Building, training, and deploying ML models traditionally requires disjointed tools and massive servers. SageMaker provides a fully managed, unified platform to do all three steps in the cloud, radically accelerating the ML lifecycle.
Tutorial Overview
In this tutorial, you will learn:
- What Amazon SageMaker is.
- The Build -> Train -> Deploy workflow.
What is Amazon SageMaker?
Amazon SageMaker is a fully managed service that provides every developer and data scientist with the ability to build, train, and deploy machine learning (ML) models quickly. SageMaker removes the heavy lifting from each step of the machine learning process.
- Build: SageMaker provides hosted Jupyter notebooks that let you explore and visualize your data stored in S3 securely.
- Train: You can select an algorithm (or write your own), and SageMaker will spin up a cluster of powerful EC2 instances (loaded with GPUs) to train the model, then immediately shut them down to save money.
- Deploy: Once the model is trained, SageMaker can instantly deploy it to a secure HTTPS endpoint where your application can begin making real-world predictions.
Exercise
?Which fully managed AWS service empowers data scientists to build, train, and deploy custom machine learning models quickly?