Using Amazon SageMaker for Computer Vision
Amazon Rekognition is a purpose-built computer vision service for 80% of everyday use cases. But what if you need more control and influence over that remaining 20%? In that case, you’d choose tooling from a lower level of AWS AI Layered Cake. These situations are precisely where Amazon SageMaker shines:
Figure 10.1 – AWS AI Layered Cake
Amazon SageMaker aims to make machine learning (ML) possible for every developer, business analyst, and data scientist. It achieves this goal through a fully managed suite of services and tools that address each ML model life cycle management step. You can start building with fully managed Juypter notebooks and train your model at scale using elastic resource pools. Suppose you need features for data labeling, data wrangling, and creating human-in-the-loop (HITL) workflows. In that case, Amazon SageMaker GroundTruth, Data Wrangler, and Augmented AI (A2I) have you...