This blog post dives into the world of AI on the edge, and how to deploy TensorFlow Lite models on edge devices. We’ll explore the challenges of managing dependencies and updates for these models, and how containerisation with Ubuntu Core and Snapcraft can streamline the process.
Let’s start by defining what TensorFlow and its Lite variant are.
TensorFlow and its sibling TensorFlow Lite
TensorFlow is a machine learning platform that implements the current best practices. It provides tools for creating ML models, running them, monitoring and improving them. TensorFlow aims to assist beginners and professionals deploying to production environments on desktop, mobile, web and cloud.
TensorFlow Lite is a library meant for running ML models on the edge or on platforms where resource constraints are greater, for example microcontrollers, embedded systems, mobile phones and so on. TF Lite is ideal when the only thing you need to do is to run an ML model. The TensorFlow Lite runtime is…
