Machine Learning Inference in IoT at the Edge – Full Lifecycle

UST Room #421

3:30 pm - 5:30 pm

Data & Machine Learning


 

How This Fits Into IoT
We will be working on IP microcontroller based IoT devices at the edge. These are the devices that will do the majority of sensor interaction in the world of IoT. We will briefly touch on using Gateways and larger devices, but the majority of the workshop will be focused on the tiny devices at the edge collecting the data.

Learning Objectives
The goal of the workshop is to show how to run ML Inference at the edge on resource-constrained devices. In addition, we will look at how to maintain the models over time through updates and how to effectively use the data you collect.

What Workshop Attendees Will Do
– Deploy ML models to the edge and run inference to collect data
– Send data back up to the cloud for further processing
– Issue firmware updates to the device with updates models

What Attendees Bring
A laptop that can talk to USB devices (Windows / OSX / Linux) and an up-to-date web browser (Firefox/Chrome)

Workshop Pre-Setup
There will be a couple of different accounts needed, this information is forthcoming.

Suggested Knowledge
Should be familiar with C or C++ or similar like languages, should familiar with machine learning concepts but not necessary. Being familiar with REST APIs would also be nice but not required. There will be a couple of different accounts needed, this information is forthcoming. There will be self-guided documentation or instructional setup videos forthcoming.

Attendees Will Receive
We will have stickers, T-Shirts, SWAG.

Further Reading Material
Blog posts forthcoming – we’re very much on the bleeding edge so this will be publicized soon.

X