feature article
Subscribe Now

Two New Ways to Program Your AI

NXP and Eta Compute Roll Out ML-At-The-edge Programming Tools

“Some people have a way with words, and other people, uh, not have way.” – Steve Martin

The scariest part of moving to a new country is learning a new language. You step off the plane and into a new land with illegible signs, strange customs, and unfamiliar culture. Which way do I go? Whom do I ask for help? Where do I even start? 

The strange world of AI/ML is terra incognita for most of us. We’ve heard rumors of its vast richness and unlimited opportunity. It certainly sounds exciting, for those willing to take the plunge. But how do you navigate its unfamiliar territory without a local guide or even a phrasebook to get you started? 

Fortunately, the ML-on-IoT landscape is populated by friendly natives willing to lend a hand. They want you to succeed, and they’re willing to donate their time to make the transition as easy as possible. Maybe, one day, you’ll become a productive member of their society. Which means you might also become a paying customer. 

Say hello (or Grüß Gott, ¡Hola!, Χαίρετε, xin chào, or こんにちは) to TensAI Flow and Glow, two new programming tools from Eta Compute and NXP, respectively. Although they’re as different as Swedish and Swahili, both exist for the same purpose: to get you conversant with machine-learning coding as painlessly as possible. 

Eta Compute, you’ll recall, is the company that produces the ECM3532 microcontroller chip, a two-headed MCU-with-DSP created especially for low-power inference at the edge. The company is convinced that uploading sensor data to a cloud server is the wrong way to do things and has dedicated its hardware and software efforts to overturning that usage model. Chips should be smart enough to collect and analyze sensor data without outside help and without draining your batteries or bank account. 

Making the hardware was tough enough, but making it easy to program may have been even tougher. TensAI Flow is the company’s new software toolchain to ease the burden. It takes high-level models created or collected by TensorFlow Lite, Edge Impulse, or ONNX, links them with its neural network zoo, and converts them to ECM35xx binaries. The software bundle also comes with libraries for common functions like image classification, motion detection, and sound processing, as well as drivers for popular sensors. TensAI Flow handles the nontrivial task of dividing up the workload between the chip’s ARM processor and its DSP companion. Finally, it produces runtime executables for FreeRTOS, the lightweight scheduler. 

TensAI Flow can’t train your neural network for you, but it significantly eases the task of getting that model onto real hardware. That’s a boon for pioneers in this largely unexplored world of ML-on-IoT programming. It provides a quick reality check for new projects, and it’s a quick-turnaround tool for iterating once you get further along. And it’s free. Some parts of the TensAI Flow package are open-sourced, but most is Eta Compute’s original work, freely licensed as long as it’s for Eta Compute hardware. 

Over on the other side of the world, chip behemoth NXP is also polishing up its ML credentials with the release of its Glow compiler for microcontrollers. Glow is a machine-learning compiler originally created at Facebook but now open-sourced and gaining industry support. Like any compiler, it’s language-specific but hardware-agnostic. A generic out-of-the-box Glow compiler will produce acceptable results (like gcc) but nothing to write home about. 

That’s where NXP comes in. The company has optimized its Glow compiler for its own line of i.MX RT microcontrollers, which – just like those at Eta Compute – are based on an ARM Cortex-M paired with a DSP core. (Weirdly, NXP doesn’t use its own CoolFlux DSP, opting instead for Tensilica’s HiFi 4 design.) The compiler accepts PyTorch, ONNX, or TensorFlow Lite input, should you decide to go that route. NXP’s Glow implementation includes CMSIS-NN libraries and other chip-specific tweaks to get the best out of your models. 

It certainly seems to work. In NXP’s benchmarks, the company showed a 2× improvement in performance just by linking in the CMSIS-NN library, and a huge 13× improvement when the compiler targets the on-chip DSP. Let’s hear it for hardware/software optimization. 

As NXP’s Director of Enabling Technologies, Markus Levy, points out, ML code is just middleware — it’s never the entire application, so performance gains mean more headroom for your “real” code. Your ML algorithms may hold your most valuable and hard-won IP, but they’re not what the customer sees. A big jump in compiler efficiency is like getting free CPU cycles to spend elsewhere. 

NXP and Eta Compute are both betting big on high-volume ML hardware, and both know that accessible software is the way to move chips. Maybe their new tools can bridge the language barrier and get more developers talking.

Leave a Reply

featured blogs
Nov 23, 2022
The current challenge in custom/mixed-signal design is to have a fast and silicon-accurate methodology. In this blog series, we are exploring the Custom IC Design Flow and Methodology stages. This methodology directly addresses the primary challenge of predictability in creat...
Nov 22, 2022
Learn how analog and mixed-signal (AMS) verification technology, which we developed as part of DARPA's POSH and ERI programs, emulates analog designs. The post What's Driving the World's First Analog and Mixed-Signal Emulation Technology? appeared first on From Silicon To So...
Nov 21, 2022
By Hossam Sarhan With the growing complexity of system-on-chip designs and technology scaling, multiple power domains are needed to optimize… ...
Nov 18, 2022
This bodacious beauty is better equipped than my car, with 360-degree collision avoidance sensors, party lights, and a backup camera, to name but a few....

featured video

Maximizing Power Savings During Chip Implementation with Dynamic Refresh of Vectors

Sponsored by Synopsys

Drive power optimization with actual workloads and continually refresh vectors at each step of chip implementation for maximum power savings.

Learn more about Energy-Efficient SoC Solutions

featured paper

How SHP in plastic packaging addresses 3 key space application design challenges

Sponsored by Texas Instruments

TI’s SHP space-qualification level provides higher thermal efficiency, a smaller footprint and increased bandwidth compared to traditional ceramic packaging. The common package and pinout between the industrial- and space-grade versions enable you to get the newest technologies into your space hardware designs as soon as the commercial-grade device is sampling, because all prototyping work on the commercial product translates directly to a drop-in space-qualified SHP product.

Click to read more

Featured Chalk Talk

Direct Drive: Getting More Juice from Your JFET

Sponsored by Mouser Electronics and UnitedSiC

In this episode of Chalk Talk, Jonathan Dodge from UnitedSiC (now part of Qorvo) and Amelia Dalton discuss how you can take full advantage of silicon carbide JFET transistors. They delve into the details of these innovative transistors including what their capacitances look like, how you can control their speed and how you can combine the benefits of a cascode and a directly driven JFET in your next design.

Click here for more information about UnitedSiC UF4C/SC 1200V Gen 4 SiC FETs