Hi, I want to run simple CNN(Convolutional Neural Network) using nrf52840 dk board using microTVM. I saw there was a guide about microtvm using Zephyr rtos.
Is that microTVM flow is working this way?
- Run TVM network using onnx/pytorch <= optimizing inference computation.
- Generate C/C++ code.
- We compile and build using Zephyr build environment and run inferencing.
If there are some missing parts, please let me know.
Thank you