Evaluating Machine Learning Models

In this section you're going to learn how to evaluate machine learning models. We'll start by learning about overfitting and underfitting and how you can partition your data into a training- and a test partition to easily detect when overfitting is happening. Then you'll learn about minibatch training, a cool trick to quickly train a model on a huge dataset without requiring a lot of computing power. 

Finally, we will look at k-fold cross validation, a neat trick to evaluate if we can safely train a model on a very small dataset.