In this chapter, we covered the training and visualization of a simple neural network using R. Here, we can change the number of neurons, the number of hidden layers, the activation functions, and so on, to determine the training of the model.

While dealing with a regression problem, the last layer is a single unit, which will give continuous values. For a classification problem, there are n terminal units, each representing the class of output with its probability. The breast cancer example had two output neurons to represent the two classes of values that are output from the neural network.

We have learned how to train, test, and evaluate a dataset using NN model. We have also learned how to visualize the NN model in R environment. We have covered the concepts like early stopping, avoiding overfitting, generalization of NN, and scaling of NN parameters.