Skip to main content

How to run ML Applications on Particle Hardware

With the release of TensorFlow Lite at Google I/O 2019, the accessible machine learning library is no longer limited to applications with access to GPUs. You can now run machine learning algorithms on microcontrollers much more easily, improving on-board inference and computation.

[Brandon Satrom] published a demo on how to run TFLite on Particle devices (tested on Photon, Argon, Boron,  and Xenon) making it possible to make predictions on live data with pre-trained models. While some of the easier computation that occurs on MCUs requires manipulating data with existing equations (mapping analog inputs to a percentage range, for instance), many applications require understanding large, complex sets of sensor data gathered in real time. It’s often more difficult to get accurate results from a simple equation.

The current method is to train ML models on specialty hardware, deploy the models on cloud infrastructure, and backhaul sensor data to the cloud for inference. By running the inference and decision-making on-board, MCUs can simply take action without backhauling any data.

He starts off by constructing a simple TGLite model for MCU execution, using mean squared error for loss and stochastic gradient descent for the optimization. After training the model on sample data, you can save the model and convert it to a C array for the MCU. On the MCU, you can load the model, TFLite libraries, and operations resolver, as well as instantiate an interpreter and tensors. From there you invoke the model on the MCU and see your results!

[Thanks dcschelt for the tip!]



from Hackaday https://ift.tt/2QVUtfY
Labels:

Comments

Post a Comment

Popular posts from this blog

Bill Gates steps down from Microsoft’s board to focus on philanthropy

In an announcement on Friday, Microsoft revealed that company co-founder Bill Gates has decided to step down from his role on its Board of Directors in order to focus on his philanthropic efforts at the Bill & Melinda Gates Foundation. This is Gate’s biggest change to his role at Microsoft since stepping down as company chairman in February 2014. According … Continue reading from SlashGear https://ift.tt/2We90Gu
Post a Comment
Read more

World Economic Forum launches Global AI Council to address governance gaps

The World Economic Forum is creating a series of councils that create policy recommendations for use of things like AI, blockchain, and precision medicine. Read More from VentureBeat http://bit.ly/2EKBjD4
Post a Comment
Read more

The Flexible Permanence of Copper Tape Circuits

Somewhere between shoving components into a breadboard temporarily and committing them to a piece of protoboard or a PCB lies the copper tape method. This flexible Manhattan-style method of circuitry formed the basis for [Bunnie Huang]’s Chibitronics startup, and has since inspired many to stop etching boards and start fetching hoards of copper tape. [Hales] hit the ground running when he learned about this method , and has made many a copper tape circuit in the last year or so. He offers several nice tips on his site that speak from experience with this method, and he’ll even show you how to easily work an SMD breakout board into the mix. Generally speaking, [Hales] prefers plywood as the substrate to paper or cardboard for durability. He starts by drawing out the circuit and planning where all the tape traces will go and how wide they need to be. Then he lays out copper traces and pads, rubs the tape against the substrate to make it adhere strongly, and reinforces joints and laps w...
Post a Comment
Read more