Machine Learning with SageMaker by Pearson

Course details

In this course, expert Nick Garner teaches you how to design and deploy scalable, cloud-based machine learning (ML) solutions on AWS. Learn about data preparation, automation, and solution management. Explore how to ingest, transform, and clean data using SageMaker Pipelines. Discover how to achieve automation by building and deploying end-to-end ML workflows. Find out how to monitor, secure, and optimize AWS-based ML systems.

This course offers hands-on AWS training that is ideal for software developers, data scientists, ML engineers, and DevOps professionals, blending theory with practical projects to help you design production-ready ML architectures, automate CI/CD pipelines for efficient deployment, and implement security best practices for cloud-based systems.

Concepts

Introduction

Machine learning with SageMaker - Introduction

Data Ingestion and Storage Basics

Module introduction
Learning objectives
Course overview
Overview of common data formats
Ingesting data with Amazon S3 and SageMaker Data Wrangler
Data ingestion demonstration
Data storage optimization and transfer in AWS

Data Transformation and Feature Engineering with SageMaker

Learning objectives
Data cleaning and preprocessing with SageMaker Data Wrangler
Data preprocessing demonstration
Feature scaling and encoding techniques
Handling missing values and outliers

Preparing Data for Modeling

Learning objectives
Validating data quality with AWS tools
Configuring data for SageMaker training jobs
Managing SageMaker Feature Store
SageMaker Feature Store demonstration

Choosing and Training Models in SageMaker

Module introduction
Learning objectives
Overview of SageMaker built-in algorithms and JumpStart models
SageMaker algorithms demonstration
Setting up and running SageMaker training jobs
SageMaker training demonstration
Hyperparameter tuning with SageMaker automatic model tuning
Hyperparameter tuning demonstration
Preventing overfitting and underfitting
Model over underfitting demonstration

Model Evaluation and Bias Detection

Learning objectives
Model evaluation metrics - Accuracy, precision, and recall
Using SageMaker Clarify for bias detection and interpretability
Comparing model performance using A B testing
Model A B testing demonstration
Managing model versions with SageMaker Model Registry
Model registry demonstration

Deploying Models with SageMaker

Module introduction
Learning objectives
Real-time inference with SageMaker endpoints
Real-time inference demonstration
Batch inference and asynchronous inference
Batch and asynchronous inference demonstration
Using SageMaker Neo for edge deployment
SageMaker edge deployment demonstration

Automating ML Workflows with SageMaker Pipelines

Learning objectives
Building and automating ML pipelines in SageMaker
SageMaker pipeline demonstration
Integrating data processing and training steps
Training and data processing in SageMaker pipelines demonstration
Triggering pipelines with EventBridge for retraining
Triggering SageMaker pipelines via EventBridge demonstration

Monitoring and Optimizing ML Solutions

Module introduction
Learning objectives
Using SageMaker Model Monitor for data drift and quality
SageMaker Model Monitor demonstration
Setting up alerts and CloudWatch dashboards
Cost optimization with auto-scaling and SageMaker Savings Plans
SageMaker auto scaling demonstration

Securing ML Models and Data in SageMaker

Learning objectives
IAM roles and permissions for SageMaker
IAM demonstration
VPC configurations for secure endpoint deployment
VPC demonstration

Summary

Machine learning with SageMaker summary