In this chapter, we will be discussing some of the more complicated parts of data science that can put some people off. The reason for this is that data science is not all fun and machine learning. Sometimes, we have to discuss and consider theoretical and mathematical paradigms and evaluate our procedures.
This chapter will explore many of these procedures step by step so that we completely and totally understand the topics. We will be discussing topics such as the following:
Cross-validation
The bias variance tradeoff
Overfitting and underfitting
Ensembling techniques
Random forests
Neural networks
These are only some of the topics to be covered. At no point do I want you to be confused. I will attempt to explain each procedure/algorithm with utmost care and with many examples and visuals.