Image Classification with Neural Networks
Up until this point, we have built models to solve both regression and classification problems on structured data with much success. The next question that comes to mind is: can we build models that can tell the difference between a dog and a cat, or a car and a plane? Today, with the aid of frameworks such as TensorFlow and PyTorch, developers can now build such ML solutions with a few lines of code.
In this chapter, we will explore the anatomy of neural networks and learn how we can apply them to building models for computer vision problems. We will start by examining what a neural network is and the architecture of a multilayer neural network. We will look at some important ideas such as forward propagation, backward propagation, optimizers, loss function, learning rate, and activation functions, and where and how they fit in.
After we build a solid base in the core fundamentals, we will build an image classifier using a custom dataset...
