Throughout this chapter, we have created and trained classification and localization models using the Oxford-IIIT-Pet dataset. We have learned how to create deep learning classifiers and localizers using transfer learning.
You have started to understand how to solve real-world problems using deep learning. You have understood how CNNs work and you know how to create a new CNN using a base model.
We have also covered the backpropagation algorithm for computing gradients. Understanding this algorithm will allow you to make wiser decisions on the architecture of models that you might want to build in the future.
In the next chapter, we will continue our deep learning journey. We will create an application that will detect and track objects with high accuracy.
