Neural Network Projects with Python

Loss functions for regression problems

It is important to understand what regression is, and how it affects the architecture of our neural network. Our task in this project is to predict taxi fares, which is a continuous variable. We can contrast this with the classification project that we did in the previous chapter, Chapter 2, Predicting Diabetes with Multilayer Perceptrons, where we designed a neural network to output a binary prediction (1 or 0), indicating whether the patient was at risk of diabetes.

Another way to think about regression and classification is that in regression, we are trying to predict the value of a continuous variable (for example, cost, time, or height), whereas in classification, we are trying to predict a class ...