Activities

This course includes presentations, hands-on labs, demonstrations, discussions, and a capstone project.

Course objectives

In this course, you will learn to:

•     Accelerate the process to prepare, build, train, deploy, and monitor ML solutions using Amazon SageMaker Studio

Intended audience

This course is intended for:

•     Experienced data scientists who are proficient in ML and deep learning fundamentals

Prerequisites

We recommend that attendees of this course have:

We recommend that all attendees of this course have:

•     Experience using ML frameworks

•     Python programming experience

•     At least 1 year of experience as a data scientist responsible for training, tuning, and deploying models

•     AWS Technical Essentials digital or classroom training

Course outline

Day 1

Module 1: Amazon SageMaker Studio Setup

·        JupyterLab Extensions in SageMaker Studio

·        Demonstration: SageMaker user interface demo

Module 2: Data Processing

·        Using SageMaker Data Wrangler for data processing

·        Hands-On Lab: Analyze and prepare data using Amazon SageMaker Data Wrangler

·        Using Amazon EMR

·        Hands-On Lab: Analyze and prepare data at scale using Amazon EMR

·        Using AWS Glue interactive sessions

·        Using SageMaker Processing with custom scripts

·        Hands-On Lab: Data processing using Amazon SageMaker Processing and SageMaker Python SDK

·        SageMaker Feature Store

·        Hands-On Lab: Feature engineering using SageMaker Feature Store

Module 3: Model Development

·        SageMaker training jobs

·        Built-in algorithms

·        Bring your own script

·        Bring your own container

·        SageMaker Experiments

·        Hands-On Lab: Using SageMaker Experiments to Track Iterations of Training and Tuning Models

Day 2

Module 3: Model Development (continued)

·        SageMaker Debugger

·        Hands-On Lab: Analyzing, Detecting, and Setting Alerts Using SageMaker Debugger

·        Automatic model tuning

·        SageMaker Autopilot: Automated ML

·        Demonstration: SageMaker Autopilot

·        Bias detection

·        Hands-On Lab: Using SageMaker Clarify for Bias and Explainability

·        SageMaker Jumpstart

Module 4: Deployment and Inference

·        SageMaker Model Registry

·        SageMaker Pipelines

·        Hands-On Lab: Using SageMaker Pipelines and SageMaker Model Registry with SageMaker Studio

·        SageMaker model inference options

·        Scaling

·        Testing strategies, performance, and optimization

·        Hands-On Lab: Inferencing with SageMaker Studio

Module 5: Monitoring

·        Amazon SageMaker Model Monitor

·        Discussion: Case study

·        Demonstration: Model Monitoring

Day 3

Module 6: Managing SageMaker Studio Resources and Updates

·        Accrued cost and shutting down

·        Updates

Capstone

·        Environment setup

·        Challenge 1: Analyze and prepare the dataset with SageMaker Data Wrangler

·        Challenge 2: Create feature groups in SageMaker Feature Store

·        Challenge 3: Perform and manage model training and tuning using SageMaker Experiments

·        (Optional) Challenge 4: Use SageMaker Debugger for training performance and model optimization

·        Challenge 5: Evaluate the model for bias using SageMaker Clarify

·        Challenge 6: Perform batch predictions using model endpoint

·        (Optional) Challenge 7: Automate full model development process using SageMaker Pipeline