Technical requirements
You can download the supplemental material for this chapter from this book’s GitHub repository at https://github.com/PacktPublishing/Production-Ready-Applied-Deep-Learning/tree/main/Chapter_10.
Before we deep dive into the individual techniques, we would like to introduce two libraries built on top of TensorFlow (TF). The first is TensorFlow Lite (TF Lite), which handles the TF model deployment on mobile, microcontrollers, and other edge devices (https://www.tensorflow.org/lite). Some of the techniques we will be describing are only available for TF Lite. The other library is called TensorFlow Model Optimization Toolkit. This library is designed to provide various optimization techniques for TF models (https://www.tensorflow.org/model_optimization).