In this book, we will primarily use the following libraries for deep learning: H2O, MXNet, and Keras. We will also use the Restricted Boltzmann Machine (RBM) package specifically for RBMs and deep belief networks (DBNs). In addition, we will conclude the book by using the ReinforcementLearning package.
In this chapter, we will install all of the previously listed packages. Each package can be used to train deep learning models in R. However, each has its particular strengths and weaknesses. We will explore the underlying architecture for each of these packages, which will help us to understand how they execute code. The packages have been created to allow R programmers to perform deep learning, with the exception of RBM and ReinforcementLearning, which are not written natively in R. This does have important implications for us to consider...
