AW235: Machine Learning Engineering on AWS™

Agenda:

• Course Introduction

• Introduction to Machine Learning (ML) on AWS™
  • Introduction to ML
  • Amazon SageMaker AI
  • Responsible ML

• Analyzing Machine Learning (ML) Challenges
  • Evaluating ML business challenges
  • ML training approaches
  • ML training algorithms

• Data Processing for Machine Learning (ML)
  • Data preparation and types
  • Exploratory data analysis
  • AWS™ storage options and choosing storage

• Data Transformation and Feature Engineering
  • Handling incorrect, duplicated, and missing data
  • Feature engineering concepts
  • Feature selection techniques
  • AWS™ data transformation services

• Lab 1: Analyze and Prepare Data with Amazon SageMaker Data Wrangler and Amazon EMR
• Lab 2: Data Processing Using SageMaker Processing and the SageMaker Python SDK

• Choosing a Modeling Approach
  • Amazon SageMaker AI built-in algorithms
  • Amazon SageMaker Autopilot
  • Selecting built-in training algorithms
  • Model selection considerations
  • ML cost considerations

• Training Machine Learning (ML) Models
  • Model training concepts
  • Training models in Amazon SageMaker AI

• Lab 3: Training a model with Amazon SageMaker AI

• Evaluating and Tuning Machine Learning (ML) models
  • Evaluating model performance
  • Techniques to reduce training time
  • Hyperparameter tuning techniques

• Lab 4: Model Tuning and Hyperparameter Optimization with Amazon SageMaker AI

• Model Deployment Strategies
  • Deployment considerations and target options
  • Deployment strategies
  • Choosing a model inference strategy
  • Container and instance types for inference

• Lab 5: Shifting Traffic

• Securing AWS™ Machine Learning (ML) Resources
  • Access control
  • Network access controls for ML resources
  • Security considerations for CI/CD pipelines

• Machine Learning Operations (MLOps) and Automated Deployment
  • Introduction to MLOps
  • Automating testing in CI/CD pipelines
  • Continuous delivery services

• Lab 6: Using Amazon SageMaker Pipelines and the Amazon SageMaker Model Registry with Amazon SageMaker Studio

• Monitoring Model Performance and Data Quality
  • Detecting drift in ML models
  • SageMaker Model Monitor
  • Monitoring for data quality and model quality
  • Automated remediation and troubleshooting

• Lab 7: Monitoring a Model for Data Drift

• Course Wrap-up

Objectives:

• Explain ML fundamentals and its applications in the AWS™ Cloud.
• Process, transform, and engineer data for ML tasks by using AWS™ services.
• Select appropriate ML algorithms and modeling approaches based on problem requirements and model interpretability.
• Design and implement scalable ML pipelines by using AWS™ services for model training, deployment, and orchestration.
• Create automated continuous integration and delivery (CI/CD) pipelines for ML workflows.
• Discuss appropriate security measures for ML resources on AWS™.
• Implement monitoring strategies for deployed ML models, including techniques for detecting data drift.

Target audience:

• DevOps Engineers
• Developers
• SysOps Engineers
• ML Engineers
• ML professionals interested in building, deploying, and operationalizing machine learning models on AWS™.

Prerequisites:

• Familiarity with basic machine learning concepts
• Working knowledge of Python programming language and common data science libraries such as NumPy, Pandas, and Scikit-learn
• Basic understanding of cloud computing concepts and familiarity with AWS™
• Experience with version control systems such as Git (beneficial but not required)

Description:

Other Info: