In the previous section, Validating performance of a classification model, we talked about validating the performance of a classification model. Now let's look at validating the performance of a regression model.
Regression models are different in that there's no binary measure of right or wrong for individual samples. Instead, you want to measure how close the prediction is to the actual value. The closer we are to the expected output, the better the model performs.
In this section, we'll discuss three methods to measure the performance of a neural network that is used for regression. We'll first talk about how to measure the performance using different error-rate functions. We'll then talk about how to use the coefficient of determination to further validate your regression model. Finally, we'll use...
