yolov5s-v3.1 ncnn 部署（一）

目录

ncnn在windows下编译

第一步：安装VS2017

第二步：下载ncnn和protobuf源码

第三步：Start the command prompt

 第三步：build protobuf library

第四步：安装Vulkan

第五步：Build ncnn library (replace with a proper path)

---

ncnn在windows下编译

参考官网：https://github.com/Tencent/ncnn/wiki/how-to-build#build-for-windows-x64-using-visual-studio-community-2017

第一步：安装VS2017

因为我的电脑本来装了VS2019，所以直接用VS2019试一下~

Download and Install Visual Studio Community 2017 from Visual Studio 2019 Community Edition – Download Latest Free Version

第二步：下载ncnn和protobuf源码

Download protobuf-3.4.0 from https://github.com/google/protobuf/archive/v3.4.0.zip

Download ncnn from github: https://github.com/Tencent/ncnn

$ git clone https://github.com/Tencent/ncnn.git
$ cd ncnn
$ git submodule update --init

第三步：Start the command prompt

官网给的是用x64 Native Tools Command Prompt for VS 2017编译，我用的是x64 Native Tools Command Prompt for VS 2019也可！这里注意使用x64进行编译不要用x86 Native Tools Command Prompt for VS 2019！！！

Start the command prompt: Start → Programs → Visual Studio 2019 → Visual Studio Tools → x64 Native Tools Command Prompt for VS 2019如下图：

 第三步：build protobuf library

cd <protobuf-root-dir>
mkdir build
cd build
cmake -G"NMake Makefiles" -DCMAKE_BUILD_TYPE=Release -DCMAKE_INSTALL_PREFIX=%cd%/install -Dprotobuf_BUILD_TESTS=OFF -Dprotobuf_MSVC_STATIC_RUNTIME=OFF ../cmake
nmake
nmake install

第四步：安装Vulkan

Download and install Vulkan SDK from LunarXchange

 正常下载安装到一个自己感觉舒服的位置上

第五步：Build ncnn library (replace with a proper path)

build ncnn 注意替换命令中的protobuf的路径为自己的protobuf的路径

cd <ncnn-root-dir>
mkdir -p build
cd build
cmake -G"NMake Makefiles" -DCMAKE_BUILD_TYPE=Release -DCMAKE_INSTALL_PREFIX=%cd%/install -DProtobuf_INCLUDE_DIR=<protobuf-root-dir>/build/install/include -DProtobuf_LIBRARIES=<protobuf-root-dir>/build/install/lib/libprotobuf.lib -DProtobuf_PROTOC_EXECUTABLE=<protobuf-root-dir>/build/install/bin/protoc.exe -DNCNN_VULKAN=ON ..
nmake
nmake install

Note: To speed up compilation process on multi core machines, configuring cmake to use jom or ninja using -G flag is recommended.

至此ncnn就编译完成了，待用

接下来去转模型pt2onnx2ncnn,转到下一篇“yolov5s-v3.1 ncnn 部署（二）”