Amazon SageMaker is a fully managed, modular service that helps developers and data scientists to build, train, and deploy machine learning models at scale. Get started with these developer resources, so you can quickly move from concept to production.
Introduction to Amazon SageMaker
Learn about the build, train, and deploy modules of Amazon SageMaker.
In this on-demand tech talk, learn to manage the complete ML workflow through a single pane of glass using Amazon SageMaker Studio. Using SageMaker Studio, you can write code, track experiments, visualize data, and perform debugging and monitoring all within a single, integrated visual interface, which significantly boosts developer productivity.
In this training course, learn how to use Amazon SageMaker to simplify the integration of machine learning into your applications. Key topics include: an overview of machine learning and problems it can help solve, using a Jupyter Notebook to train a model based on Amazon SageMaker’s built-in algorithms and, using Amazon SageMaker to publish the validated model. You will finish the course by building a serverless application that integrates with the Amazon SageMaker published endpoint.
In this training course, you will learn how to implement a machine learning pipeline using Amazon SageMaker and Amazon SageMaker Ground Truth. First you will create a labeled dataset, then you’ll create a training job to train your object detection model, and finally you will use Amazon SageMaker to create and update your model.
Secure and compliant ML workflows with Amazon SageMaker
Ever wondered how to build a secure and compliant end-to-end ML workflow for Financial Services? Check out this video demonstration, where we address the common patterns and requirements required by highly regulated industries for their use cases with secure machine learning.
Build machine learning models
Use the Build module of SageMaker to collect and prepare training data, access pre-built notebooks, and leverage the built-in, high performance algorithms.
Learn to build an ML model with the steps and resources outlined in this guide.
In this video, learn all about the fully-managed notebook instances with Amazon SageMaker.
Read this blog and learn to use common workflows using Amazon SageMaker notebook instances.
Acccess a rich repository of SageMaker notebooks, on GitHub.
Utilize algorithms built into Amazon SageMaker that are faster and cheaper than popular alternatives.
In this video, learn about the high-performance algorithms, built-in with Amazon SageMaker.
Train and tune machine learning models
Use the Train module to set up training environments with one click and optimize your model using automatic module tuning
Read an overview of how to train machine learning models using Amazon SageMaker.
In this video, learn how to train and tune your machine learning models to the highest accuracy with Amazon SageMaker.
Organize and track your training iterations efficiently with Amazon SageMaker Experiments. Training an ML model typically entails many iterations to isolate and measure the impact of changing data sets, algorithm versions, and model parameters. SageMaker Experiments helps you manage these iterations by automatically capturing the input parameters, configurations, and results, and identify the best performing experiment.
Try these examples of using hyperparameter tuning across different algorithms and deep learning frameworks.
Learn how to save up to 90% in training costs, using Amazon EC2 Spot instances with Managed Spot Training. Spot instances are space compute capacity and training jobs are automatically run when the spare capacity becomes available. Training runs are made resilient to interruptions caused by changes in capacity, allowing you to save cost when you have flexibility with when to run training jobs.
In this on-demand tech talk, learn how to use Amazon SageMaker Experiments and how Amazon SageMaker Debugger improves model quality through better model training and tuning. You will see how to manage iterations by automatically capturing the input parameters, configurations, and results and automatically capturing real-time metrics during training such as training and validation and confusion matrices.
Deploy machine learning models
Use the Deploy module to deploy your machine learning models to production with a single click.
Follow the step-by-step guide to deploy machine learning models on the highest performing infrastructure.
In this video, learn how to deploy your ML models to production on the most scalable infrastructure.
Follow the examples on GitHub to use Amazon SageMaker and AWS Step Functions to automate the building, training, and deploying of custom machine learning models.
Learn to use the deployment capabilities of SageMaker including A/B testing and Auto Scaling, delivering high performance and high availability for your machine learning models.
In this blog, learn how to maintain quality of your machine learning models in production, when changes such as concept drift occur, using Amazon SageMaker Model Monitor. You can even get alerted when data quality issues appear, for you to take the required action.
In this blog, learn how you can build, train, and deploy fastai models into Amazon SageMaker training and hosting by using the Amazon SageMaker Python SDK and a PyTorch base image. You can avoid the extra steps of building your own container.
Use APIs tailored to your programming language or platform to make it easy to use Amazon SageMaker in your applications.
What’s New announcements are high-level summaries of launches and feature updates. Read Amazon SageMaker specific updates and other AWS announcements.