@Crocodile: thanks for your answer. Is there any way of doing an inference on a neural network on the RV1103 using the neural accelerator?

Hi, I am trying to run a python script to do inference on a neural network model.
I have created the .rknn model file on a PC and transferred it to the Luckfox, with a test image.

But I can't manage to use both rknnlite and pillow. Where can I find installation instructions?
To use the NPU, I need ...

Thanks for your answer.
Digging into the folders, I found the CMSIS DSP library here:
sysdrv/source/mcu/rt-thread/components/CMSIS/Include
But in the arm_math.h file (also here https://github.com/LuckfoxTECH/luckfox-pico/blob/a4230afbf485a5eec4fd17ce59a1e9ff2310dc2b/sysdrv/source/mcu/rt-thread ...

I am working on a neural network in C++ on the Pico Mini B.
I'd like to accelerate it by calling the dotproduct functions of the CMSIS DSP library.

But the
#include "arm_math.h"
line calls a compilation error:
./dnn/include/kernels/macs.hpp:4:10: fatal error: arm_math.h: No such file or ...

Thanks for your answer.
I tried that, but here is what I got:

Building dependency tree... Done
Reading state information... Done
E: Unable to locate package python3-periphery

Edit: I found the explanation in this video
https://www.youtube.com/watch?v=2EgyVn7HRHs

Just download the library for ...

Hi, I have the same problem with a Pico Mini B. How can I install the libraries?
Thanks for your help.

Hi, I have the same problem on my Pico Mini B.
I have tried your fix, but it doesn't work, as the Pico B does not connect to internet. Here is what I get:

bash luckfox-gcc-fix.sh
--2023-09-20 01:16:29-- http://http.us.debian.org/debian/pool/main/g/glibc/libc6-dev_2.40-4_armhf.deb
Resolving http.us ...

Can anyone help me, please?

Thanks, it should work, but doesn't.
I'll try again later on a Linux computer.

But the easiest way would be (TMHA) to burn an image with another software

Thank you. I understand that, but as I can't use the burning software, I expected that someone could write the files on an empty SD card and make an image of it, then put it somewhere where I can download it to finally burn it on my SD card (using a different software).