Part 2. Practical application


In part 2, you’ll go beyond a basic ML workflow to look at how to extract features from text, images, and time-series data to improve the accuracy of models even further, and to scale your ML system to larger data volumes. In addition, you’ll go through three full example chapters to see everything in action.

In chapter 6, our first full example chapter, you’ll try to predict the tipping behavior of NYC taxis.

In chapter 7, you’ll look at advanced feature-engineering processes that allow you to extract value out of natural language text, images, and time series data. A lot of modern ML and artificial intelligence applications are based on these techniques.

In chapter 8, you’ll use this advanced feature-engineering knowledge in another full example: predicting the sentiment of online movie reviews.

In chapter 9, you’ll learn techniques for scaling ML systems to larger volumes of data, higher prediction throughput, and lower prediction latency. These are all important aspects of many modern ML deployments.

In chapter 10, you’ll walk through a full example of building a model—on large amounts of data—that predicts online digital display advertisement clicks.