Practical Data Science with Amazon SageMaker

This intermediate-level course provides practical experience in applying machine learning techniques using Amazon SageMaker. Participants work through a real-world customer churn use case, covering the key stages of a typical data science workflow from dataset preparation and analysis to model training, evaluation, tuning, and deployment.

The course focuses on using Amazon SageMaker to prepare datasets, analyze and visualize data, train and evaluate machine learning models, perform automatic hyperparameter tuning, and prepare models for production. Participants also explore deployment strategies, auto scaling, the relative cost of classification errors, and additional SageMaker capabilities including notebooks in a VPC, batch transforms, Ground Truth, and Neo. The current AWS classroom-training catalogue lists this course as Intermediate and 1 day.

• Prepare datasets for machine learning training
• Analyze and visualize datasets to understand relationships between features and target variables
• Apply data cleaning techniques to prepare datasets for modeling
• Train and evaluate machine learning models using Amazon SageMaker
• Apply XGBoost for machine learning model development with SageMaker
• Configure and perform automatic hyperparameter tuning
• Deploy machine learning models to Amazon SageMaker endpoints
• Apply A/B deployment and auto scaling concepts for production readiness
• Analyze the relative cost of different types of model errors
• Explore Amazon SageMaker architecture and features for machine learning workflows

• Familiarity with Python programming language
• Basic understanding of Machine Learning