Use and install RkNN toolkit Lite2 on itop-3568 development board NPU

1 Type the following command to install Python3.7、pip3 、gcc

apt-get install python3 python3-dev python3-pip gcc

2 Installation dependent , Enter the following command ： notes ：Toolkit-lite2 It doesn't depend on opencv-python,

But this module is needed in the example , Therefore, it is also installed during installation . If you don't use image correlation

To deal with , You may not install the dependent Library

apt-get install -y python3-opencv

apt-get install -y python3-numpy

3 Install dependency packages setuptools, Enter the following command ：

pip install -i https://pypi.tuna.tsinghua.edu.cn/simple setuptools

4 Enter the following command to upgrade pip

python3 -m pip install --upgrade pip

5 Download the network disk data “iTOP-3568 Development board \02_【iTOP-RK3568 Development board 】 Development of information

\11_NPU Use supporting materials \04_RKNN-Toolkit-lite2 Tools ” in rknn_toolkit_lite2.tar.gz Pressure

Shrink package , stay windows Decompress , And then through U Disk copy

rknn_toolkit_lite2/packages/rknn_toolkit_lite2-1.3.0-cp37-cp37m-linux_aarch64.whl

On the development board . As shown in the figure below ：

6 install Toolkit-lite2,Debian10 ARM64 with python3.7

pip3 install rknn_toolkit_lite2-1.3.0-cp37-cp37m-linux_aarch64.whl

More content can be paid to official account. ： Beijing Xunwei