Amazon SageMaker Studio for Data Scientists

Course Modules

Day 1:

1. Module 1: Amazon SageMaker Studio Setup

   • Introduction to SageMaker Studio's interface and features.
   • Setting up and navigating the environment.

2. Module 2: Data Processing

   • Utilizing SageMaker Data Wrangler for data preparation.
   • Hands-On Lab: Data analysis and preparation with Data Wrangler.
   • Processing data at scale using Amazon EMR.
   • Hands-On Lab: Data processing with Amazon EMR.
   • Employing AWS Glue for interactive data sessions.
   • Using SageMaker Processing with custom scripts.
   • Hands-On Lab: Data processing with SageMaker Processing and Python SDK.
   • Feature engineering using SageMaker Feature Store.
   • Hands-On Lab: Implementing feature engineering with Feature Store.

3. Module 3: Model Development

   • Creating and managing SageMaker training jobs.
   • Leveraging built-in algorithms and custom scripts.
   • Utilizing custom containers for model training.
   • Tracking experiments with SageMaker Experiments.
   • Hands-On Lab: Managing training iterations with SageMaker Experiments.

Day 2:

1. Continuation of Module 3: Model Development

   • Analyzing training performance with SageMaker Debugger.
   • Hands-On Lab: Performance analysis using SageMaker Debugger.
   • Automating model tuning processes.
   • Exploring SageMaker Autopilot for automated ML.
   • Detecting biases with SageMaker Clarify.
   • Hands-On Lab: Evaluating models for bias and explainability.
   • Utilizing SageMaker JumpStart for quick model deployment.

2. Module 4: Deployment and Inference

   • Managing models with SageMaker Model Registry.
   • Automating workflows using SageMaker Pipelines.
   • Hands-On Lab: Implementing Pipelines and Model Registry.
   • Exploring various inference options with SageMaker.
   • Scaling models for production environments.
   • Hands-On Lab: Performing inferences within SageMaker Studio.

3. Module 5: Monitoring

   • Setting up model monitoring with SageMaker Model Monitor.
   • Case studies on effective model monitoring.
   • Demonstration: Real-time model monitoring techniques.

Day 3:

1. Module 6: Managing SageMaker Studio Resources and Updates

   • Managing costs and resource utilization within SageMaker Studio.
   • Procedures for updating and maintaining the environment.

2. Capstone Project

   • Hands-On Lab: End-to-end ML project using SageMaker Studio.
   • Challenges covering data analysis, feature engineering, model training, bias evaluation, and deployment.
   • Optional Challenge: Automating the ML workflow with SageMaker Pipelines.