This chapter covers
- Getting familiar with the end-to-end pipeline for conducting an ML project
- Preparing data for ML models (data collection and preprocessing)
- Generating and selecting features to enhance the performance of the ML algorithm
- Building up linear regression and decision tree models
- Fine-tuning an ML model with grid search
Now that the first chapter has set the scene, it’s time to get familiar with the basic concepts of ML and AutoML. Because AutoML is grounded in ML, learning the fundamentals of ML will help you better understand and make use of AutoML techniques. This is especially the case when it comes to designing the search space in an AutoML algorithm, which characterizes the ML components to be used and the ranges of their hyperparameters. In this chapter, we will walk through a concrete example of solving an ML problem. This will help you gain a deeper understanding of the overall process of building up an ML pipeline, especially if you have little experience working on ML projects. You will also learn a naive way of tuning the hyperparameters of an ML model. This can be thought of as one of the simplest applications of AutoML, showing how it can help you find a better ML solution. More advanced AutoML tasks and solutions will be introduced in the second part of the book.