Summary

In this chapter, the reader has seen some advanced deep learning techniques. First, we looked at some image classification models and looked at some historical models. Next, we loaded an existing model with pre-trained weights into R and used it to classify a new image. We looked at transfer learning, which allows us to reuse an existing model as a base on which to build a deep learning model for new data. We built an image classifier model that could train on image files. This model also showed us how to use data augmentation and callbacks, which are used in many deep learning models. Finally, we demonstrated how we can build a model in R and create a REST endpoint for a prediction API that can be used from other applications or across the web.

We have come to the end of the book, and I really hope it was useful to you. R is a great language for data science and I believe it is easier to use and allows you to develop machine learning prototypes faster than the main alternative, Python...