Overview
Amazon Web Services (AWS) SageMaker is a cloud machine learning service that lets developers build, train, and deploy machine learning models quickly at any scale.
This instructor-led, live training (online or onsite) is aimed at data scientists and developers who wish to create and train machine learning models for deployment into production-ready hosting environments.
By the end of this training, participants will be able to:
- Use notebook instances to prepare and upload data for training.
- Train machine learning models using training datasets.
- Deploy trained models to an endpoint to create predictions.
Format of the Course
- Interactive lecture and discussion.
- Lots of exercises and practice.
- Hands-on implementation in a live-lab environment.
Course Customization Options
- To request a customized training for this course, please contact us to arrange.
Requirements
- Experience with application development
- Familiarity with Amazon Web Services (AWS) Console
Audience
- Data scientists
- Developers
Course Outline
Introduction
- Understanding machine learning with SageMaker
- Machine learning algorithms
Overview of AWS SageMaker Features
- AWS and cloud computing
- Models development
Setting up AWS SageMaker
- Creating an AWS account
- IAM admin user and group
Familiarizing with SageMaker Studio
- UI overview
- Studio notebooks
Preparing Data Using Jupyter Notebooks
- Notebooks and libraries
- Creating a notebook instance
Training a Model with SageMaker
- Training jobs and algorithms
- Data and model parallel trainings
- Post-training bias analysis
Deploying a Model in SageMaker
- Model registry and model monitor
- Compiling and deploying models with Neo
- Evaluating model performance
Cleaning Up Resources
- Deleting endpoints
- Deleting notebook instances
Troubleshooting
Summary and Conclusion