The files that Edge Impulse generates for model inference, which inference engine Edge Impulse use under the hood?

#1

The files that Edge Impulse generates for model inference, which inference engine Edge Impulse use under the hood?

#2

It depends on the hardware. For general purpose MCUs we typically use EON Compiler w/ TFLite Micro kernels (incl. HW optimization, e.g. via CMSIS-NN), on Linux if you run on CPU we use TensorFlow Lite. But then for accelerators we’ll use a wide variety of other runtimes, e.g. hardcoded network in silicon for Syntiant, custom SNN based inference engine for Brainchip Akida, DRP-AI for Renesas RZV2L, etc.

1 Like