Get the latest tech news
Implementing neural networks on the "3 cent" 8-bit microcontroller
Bouyed by the surprisingly good performance of neural networks with quantization aware training on the CH32V003, I wondered how far this can be pushed. How much can we compress a neural network whi…
The smallest device of the portfolio, the PMS150C, sports 1024 13-bit word one-time-programmable memory and 64 bytes of ram, more than an order of magnitude smaller than the CH32V003. The plot above shows the result of my hyperparameter exploration experiments, comparing models with different configurations of weights and quantization levels from 1 to 4 bit for input images of 8×8 and 16×16. A lot of memory footprint and processing overhead is usually spent on implementing flexible inference engines, that can accomodate a wide range of operators and model structures.
Or read this on Hacker News
