December 2019
Intermediate to advanced
528 pages
11h 19m
English
This is a huge topic, and far too big to cover in any kind of sensible depth in one section. However, it's worth covering, briefly, what machine learning is, and what it is not.
In this chapter, we'll be taking an image of a face and uploading it into a machine learning service. This service will provide details regarding whether it thinks that face is happy, sad, angry, or anything else. It's important to try and not anthropomorphise this process. The algorithm will analyze the face and return percentages of certainty; however, this is based on it seeing many, many similar images and being told whether the image it was identifying was, say, happy. This means that if I start to feed images of sad people in, but tell the machine ...