nano pi neo + ncnn环境搭建

环境

nano pi neo(arm v7) +win10（ssh）

• ARMv7 (and below) is 32-bit.

which bash
/bin/bash
file /bin/bash
/bin/bash: ELF 32-bit LSB executable, ARM, EABI5 version 1 (SYSV), dynamically linked, interpreter /lib/ld-linux-armhf.so.3, for GNU/Linux 3.2.0, BuildID[sha1]=13d7757f1c65cc7166875b5a0a4c22359b3fb426, stripped

nano pi neo 的调试

• 参考文章 NanoPi-NEO——初体验
• nano pi neo 官方文档

CPU: Allwinner H3, Quad-core Cortex-A7 Up to 1.2GHz
DDR3 RAM: 256MB/512MB
Connectivity: 10/100M Ethernet
USB Host: Type-A x 1, 2.54 mm pin x 2
MicroSD Slot x 1
MicroUSB: for data transmission and power input
Debug Serial Port: 4Pin, 2.54 mm pitch pin header
GPIO: 2.54mm pitch 36pin. It includes UART, SPI, I2C, IO etc
PCB Dimension: 40 x 40 mm
Power Supply: DC 5V/2A
OS/Software: u-boot and UbuntuCore

• ubuntu 基础

ncnn 编译

• NCNN介绍与arm-v7编译
• 参考ncnn wiki
• 参考ncnn-pc-csdn
• arm v7 Tensorflow Lite
• protobuf官方文档

# install g++ cmake protobuf
sudo apt-get install autoconf automake libtool curl make g++ unzip
# build  protobuf
git clone https://github.com/protocolbuffers/protobuf
# 或者 wget https://github.com/protocolbuffers/protobuf/releases/download/v3.13.0/protobuf-cpp-3.13.0.tar.gz 
# tar -xzvf xxx.tar.gz
cd protobuf
git submodule update --init --recursive
./autogen.sh
./configure
make
make check # 失败后 make clean 重新make
sudo make install
sudo ldconfig # refresh shared library cache.
#------------
# ARM V7 make出错，算了，直接安装ncnn

# -- Could NOT find Protobuf (missing: PROTOBUF_LIBRARY PROTOBUF_INCLUDE_DIR) CMake Warning at tools/onnx/CMakeLists.txt:16 (message): Protobuf not found, onnx model convert tool won't be built
参考ncnn wiki
多种平台build

Build for Raspberry Pi 3
install g++ cmake protobuf
cd ncnn-root-dir
mkdir -p build
cd build
cmake -DCMAKE_TOOLCHAIN_FILE=…/toolchains/pi3.toolchain.cmake -DPI3=ON …
make -j4
make install
pick build/install folder for further usage