If you have recently started working with supervised machine learning, you will have built a few models to make predictions with data. This may come with varying degrees of success. In this course, we will examine some of the problems you may have encountered when building your models and how these can be addressed.
First, we will examine the problems of underfitting, overfitting and multicollinearity and learn how to spot them.
We will then look at a range of ways to evaluate both classification and regression models, allowing you to choose a method that best suits your task.
Finally, we will take a look at a number of ways to improve your models, including feature selection, cross-validation, hyper-parameter tuning and regularization.
