Chapter 5: Training Computer Vision Models
In the previous chapter, you learned how to use SageMaker's built-in algorithms for traditional machine learning problems including classification, regression, and anomaly detection. We saw that these algorithms work well on tabular data, such as CSV files. However, they are not well suited for image datasets, and they generally perform very poorly on computer vision (CV) tasks.
For a few years now, CV has taken the world by storm, and not a month goes by without a new breakthrough in extracting patterns from images and videos. In this chapter,you will learn about three built-in algorithms designed specifically for CV tasks.We'll discuss the types of problems that you can solve with them. We'll also spend a lot of time explaining how to prepare image datasets, as this crucial topic is often inexplicably overlooked. Of course, we'll train and deploy models too.
This chapter covers the following topics:
- Discovering...
