Part 2: Going Faster
In this part, you will learn about the main techniques and approaches that can be used in PyTorch to accelerate the training process of deep learning models. First, you will learn how to compile a model by using the Compile API. After that, you will learn how to use and configure specialized libraries to optimize the training process on CPUs. Then, you will learn how to build an efficient data pipeline to keep the GPU busy all the time. Also, you will learn how to simplify the model by applying pruning and compression techniques. Finally, you will learn how to adopt automatic mixed precision to reduce computing time and memory consumption.
This part has the following chapters:
- Chapter 3, Compiling the Model
- Chapter 4, Using Specialized Libraries
- Chapter 5, Building an Efficient Data Pipeline
- Chapter 6, Simplifying the Model
- Chapter 7, Adopting Mixed Precision