Deep learning is probably the hottest technology in data science right now, and R is one of the most popular data science languages. However, R is not considered as an option for deep learning by many people, which is a shame, as R is a wonderful language for data science. This book shows that R is a viable option for deep learning, because it supports libraries such as MXNet and Keras.
When I decided to write this book, I had numerous goals. First, I wanted to show how to apply deep learning to various tasks, and not just to computer vision and natural language processing. This book covers those topics, but it also shows how to use deep learning for prediction, regression, anomaly detection, and recommendation systems. The second goal was to look at topics in deep learning that are not covered well elsewhere; for example, interpretability with LIME, deploying models, and using the cloud for deep learning. The last goal was to give an overall view of deep learning and not just provide machine learning code. I think I achieved this by discussing topics such as how to create datasets from raw data, how to benchmark models against each other, how to manage data when model building, and how to deploy your models. My hope is that by the end of this book, you will also be convinced that R is a valid choice for use in deep learning.
