In this chapter, we will explore the different modules of deep learning architectures that are used to solve real-world problems. In the previous chapter, we used low-level operations of PyTorch to build modules such as a network architecture, a loss function, and an optimizer. In this chapter, we will explore some of the important components of neural networks required to solve real-world problems, along with how PyTorch abstracts away a lot of complexity by providing a lot of high-level functions. Towards the end of the chapter, we will build algorithms that solve real-world problems such as regression, binary classification, and multi-class classification.

In this chapter, we will go through following topics:

• Deep dive into the various building blocks of neural networks
• Exploring higher-level functionalities in PyTorch to build deep learning architectures
• Applying deep learning to a real-world image classification problem