In this chapter, we will explore what was one of the most popular mainstream applications of deep learning in 2017—style transfer. We begin by introducing the concepts of style transfer and then its faster alternative, appropriately named fast neural style transfer. Similar to other chapters, we will provide the intuition behind the models (rather than granular details) and, in doing so, you will gain a deeper understanding and appreciation for the potential of deep learning algorithms. Unlike previous chapters, this chapter will focus more on the steps involved in getting the model working on iOS rather than building up the application, in order to keep it concise. 

By the end of this chapter you will have achieved the following:

• Gained an intuitive understanding of how style transfer works
• Gained hands-on experience of working with the Core ML Tools Python package and custom layers to get Keras models working in Core ML

Let's get started by introducing style transfer and building our understanding of how it works.