The Building Blocks of Machine Learning and Deep Learning

This chapter takes a look at the three main categories of machine learning and then moves on to explore how the machine learning models can be evaluated. The various metrics commonly used are explained. After that, the important topic of data preprocessing is briefly addressed followed by standard methods of evaluating machine learning models. One of the reasons why most models fail to perform on unseen data is due the problem of overfitting. We take a look at this problem and outline some of the strategies that can be applied in order to overcome it. The next topic covered in this chapter is a discussion of the workflow for machine learning or deep learning. This workflow gives the reader clear steps to successfully evaluate the performance of deep learning models given a set of data. Finally, some self-evaluation exercises are given to emphasize the key takeaways from the chapter. We also provide a list of references for further reading.