search

mit-han-lab / tinyengine

[NeurIPS 2020] MCUNet: Tiny Deep Learning on IoT Devices; [NeurIPS 2021] MCUNetV2: Memory-Efficient Patch-based Inference for Tiny Deep Learning; [NeurIPS 2022] MCUNetV3: On-Device Training Under 256KB Memory
https://mcunet.mit.edu
MIT License
792 stars 130 forks source link

Demo without LCD/Camera #33

Closed irvingzhang0512 closed 1 year ago

irvingzhang0512 commented 1 year ago

Thanks for the great work!

For now, the demos are all strongly related to the camera and LCD. I think a demo without camera/lcd, and read image from head file and print output results will be very helpful.

BTW: I'm working on this demo in spare time, but I'm familiar with stm32... Don't know how much time I need to finish this

meenchen commented 1 year ago

Hi @irvingzhang0512,

Thanks for reaching out. You can use the TESTTENSOR macro to remove the dependency on LCD/Camera. https://github.com/mit-han-lab/tinyengine/blob/03639563ebf6538fff557515e31667fca6448cd3/tutorial/inference/Src/main.cpp#L80

To set the input image, please refer to the following code block https://github.com/mit-han-lab/tinyengine/blob/03639563ebf6538fff557515e31667fca6448cd3/tutorial/inference/Src/main.cpp#L111

Hope this could help you achieve your goal. Let me know if you have further questions.

meenchen commented 1 year ago

Close due to inactivity. Feel free to reopen.