Giters

tum-ei-eda / mlonmcu

Tool for the deployment and analysis of TinyML applications on TFLM and MicroTVM backends

Geek Repo:Geek Repo

Github PK Tool:Github PK Tool

Deploy Examples/Demos on actual Hardware

PhilippvK opened this issue · comments

commented

The current focus of mlonmcu target software is to benchmark inference performance and validate model outputs for predefined input data.

As support for real hardware targets (e.g. esp32, esp32) was recently introduced it would be great to also use the on-board peripherals (i.e. microphone/MEMS sensors) in the deployed target software (if feasible).

The following points have to be consider:

  • How to switch between benchmark/demo mode? - Command line flag? Optional run stage?
  • How to design abstraction interface for target-specific code in per-model mlif_overwrite code? Multiple files? Compiler macros? (#if defined(ESPIDF_PLATFORM) && defined(RUN_DEMO) ....)
  • As the demo should not end after a single inference we would need a way to keep the model running (e.g. --num=inf) - Should mlonmcu stop after flashing or also monitor the hardware?
commented

This got partly implemented by the esp-idf platform templates. Closing for now