Embedded Learning Library
Microsoft has released the Embedded Learning Library, offering developers a pre-trained image recognition model for Raspberry Pi and other developer boards.
Early preview of Embedded Learning Library (ELL) is part of Microsoft’s effort to miniaturize machine-learning software for a range of extremely low-powered chips on devices that aren’t connected to cloud.
A team at the Microsoft Research lab is working on compressing its machine learning models to work on the Cortex-M0, an ARM processor no bigger than breadcrumb.
The aim is to push machine learning to devices without internet, such as brain implants. Microsoft’s new art feature for its Pix iPhone photo app uses AI on the device. But, the plan is to enable it on much less powerful chips, such as a brain implant, which might need to work without a network connection.
Its current compression efforts resulted in machine learning models 10 to 100 times smaller. But, to get it running on a Cortex M0, the models need to be 1,000 to 10,000 times smaller.
Today, ELL is available for the relatively powerful and large Raspberry Pi, Arduinos, BBC’s microbit and other microcontrollers. ELL devices relies on compressed machine learning models that trained for cloud. The devices work on Cortex-M0 training algorithms that tuned for specific scenarios.
The smallest device the researchers have tested is the single-board computer, Arduino Uno, which has 2 kilobytes of RAM.
A group of Microsoft researchers trained a computer vision model to deal with a squirrel problem in his yard. Researchers deployed the model on a Raspberry Pi 3 hooked up with a webcam, which switches on sprinkler system when it detects a squirrel.
Microsoft offered instructions on GitHub for makers started with similar system, which recognizes objects and prints a label.