Letter #15 - In which we remember to be good.
“Be good, or at least better than the other guys.”
Remember me when you’re the one who’s silver screen
Remember me when you’re the one you always dreamed
Remember me when everyone’s noses start to bleed
Remember me, special needs
For example, the MobileNetV2 image classification network takes 22 million ops (each multiply-add is two ops) in its smallest configuration. If I know that a particular system takes 5 picojoules to execute a single op, then it will take (5 picojoules * 22,000,000) = 110 microjoules of energy to execute. If we’re analyzing one frame per second, then that’s only 110 microwatts, which a coin battery could sustain continuously for nearly a year. These numbers are well within what’s possible with DSPs available now, and I’m hopeful we’ll see the efficiency continue to increase
I could make a joke here about creating a blockchain-based AI startup that also makes use of IoT :)
DeepLens is essentially a small Ubuntu- and Intel Atom-based computer with a built-in camera that’s powerful enough to easily run and evaluate visual machine learning models. In total, DeepLens offers about 106 GFLOPS of performance.
Not as tiny as Peter Warden hopes for, but still a step in the right direction. They’re taking preorders, you know.
Not sure what you just read? Take a look at this post.
Enjoyed it? Subscribe to my tiny newsletter and get new issues straight to your inbox. Every week, on Tuesdays.