Practising Supervised Learning:
Deep Neural Networks

It’s your turn

1. Choose a task sheet. Either work with the iris data and predict categorical values or choose airquality dataset to predict numerical values
2. Open the task sheet in RStudio
3. Fill the gaps to train a deep neural network
   If you want to read what a function (e.g. dnn) does use ? to access its documentation (i.e. ?dnn)
4. Don’t be shy to play with the hyperparameters. What happens when you adjust the learning rate? Does the number of epochs have an influence? What’s with the overall architecture of your network?