Deployment of Deep Learning Networks

The surge on the Internet and data has led to advanced deep learning systems and hence the book also presents techniques for Internet of Things (IoT) in association with deep learning networks. This section discusses and reveals the computing infrastructure that sits on the edge of a network. More importantly in this section, the chapter reveals the best deployment of deep learning network on IoT edge devices and reveals the benefits of the implementation. The core areas addressed here are, how to reduce the latency, enhance the security and communicate with less bandwidth by deploying deep learning networks. Further, the chapter demonstrate and details a comprehensive way to setup, install, compile, run, test and deploy different IoT edge devices. Through this chapter readers also understand and gain strong learning in event data collection, flow data collection, vulnerability assessment, network analysis, packet inspection, android deployment diagnosis, neural data communication with android services. At the higher side, in this section the book presents how to setup and run the IBM Watson Visual Recognition Service in android and application services. Further, AI model pruning and Optimization, AI model Quantizer and Edge compilers are also discussed. The chapter enumerates case studies on agriculture connected to IoT and DL networks for reader understanding.