AWS · MLA-C02 · BETA
Choosing a modeling approach
MLA-C02 · Lesson 7 of 23 · 54:51
YouTube loads only after you press play. This chapter opens at its timestamp in the full course. Watch on YouTube ↗
In this lesson
Lesson 7 of the free MLA-C02 beta series (AWS Certified Machine Learning Engineer – Associate, domain 2, model development). In this lesson: the AWS AI services that solve a problem without training, build or buy, SageMaker AI built-in algorithms and common libraries, and the tradeoffs between performance, training time, and cost. With three quick checks and a recap, and AnyCompany Mobility, a car sharing service, as the running example.
Preview the topics
- At sign-up, the app must read the name, birth date, and license number from photos of US driver licenses. The team has no labeled data and no ML experts. What should the ML engineer use?
- The team must forecast car demand for each neighborhood, hour by hour, across hundreds of related neighborhoods, with a built-in algorithm. Which one fits?
- Regulators require AnyCompany Mobility to explain every fraud decision to customers in plain terms, and the current deep network is only slightly more accurate than simpler models. What should the ML engineer choose?