AWS · MLA-C02 · BETA
Transformations, features, and Feature Store
MLA-C02 · Lesson 3 of 23 · 17:54
YouTube loads only after you press play. This chapter opens at its timestamp in the full course. Watch on YouTube ↗
In this lesson
Lesson 3 of the free MLA-C02 beta series (AWS Certified Machine Learning Engineer – Associate, domain 1, data preparation). In this lesson: the tools that transform data (AWS Glue, Data Wrangler in SageMaker Canvas, EMR), scaling, transforming, and binning values, splitting and encoding features, and SageMaker Feature Store with feature groups, online and offline stores, and ingestion. With three quick checks and a recap, and AnyCompany Mobility, a car sharing service, as the running example.
Preview the topics
- Business analysts must clean and reshape the trips dataset every week, without writing code, and want to reuse the same steps each time. What should they use?
- Trip distances are heavily skewed: most trips are short, but a few cover hundreds of kilometers and dominate a linear model. Which transformation should the ML engineer apply?
- The fraud model needs the latest spending features of a customer during each payment, within milliseconds, and data scientists need the full history for retraining. How should the feature group be configured?