Caffe + CUDA7.5 + CuDNNv3 + OpenCV3.0 in UbuntuServer14.04配置

基本依赖项

sudo apt-get install libprotobuf-dev libleveldb-dev libsnappy-dev libopencv-dev libhdf5-serial-dev protobuf-compiler sudo apt-get install --no-install-recommends libboost-all-dev

1 2	sudo apt-get install libprotobuf-dev libleveldb-dev libsnappy-dev libopencv-dev libhdf5-serial-dev protobuf-compiler   sudo apt-get install --no-install-recommends libboost-all-dev

NVIDIA驱动

sudo apt-get install nvidia-current

1	sudo apt-get install nvidia-current

CUDA

先安装内核头文件：

sudo apt-get install linux-headers-$(uname -r)

1	sudo apt-get install linux-headers-$(uname -r)

安装cuda(官网下载deb文件）：

sudo dpkg -i cuda-repo-<distro>_<version>_<architecture>.deb sudo apt-get update sudo apt-get install cuda

1 2 3

sudo dpkg -i cuda-repo-<distro>_<version>_<architecture>.deb   sudo apt-get update   sudo apt-get install cuda

环境变量：

export PATH=/usr/local/cuda-7.5/bin:$PATH export LD_LIBRARY_PATH=/usr/local/cuda-7.5/lib64:$LD_LIBRARY_PATH

1 2	export PATH=/usr/local/cuda-7.5/bin:$PATH   export LD_LIBRARY_PATH=/usr/local/cuda-7.5/lib64:$LD_LIBRARY_PATH

测试：

cuda-install-samples-7.5.sh ~ cd ~/NVIDIA_CUDA-7.5_Samples cd 1_Utilities/deviceQuery make

1 2 3 4

cuda-install-samples-7.5.sh  ~   cd ~/NVIDIA_CUDA-7.5_Samples   cd 1_Utilities/deviceQuery   make

执行deviceQuery，如果成功结尾会是Result = PASS

cuda环境配置：

sudo nano /etc/ld.so.conf.d/cuda.conf /usr/local/cuda/lib64 /lib

1 2 3

sudo nano /etc/ld.so.conf.d/cuda.conf       /usr/local/cuda/lib64     /lib

完成lib文件的链接操作，执行：

sudo ldconfig -v

1	sudo ldconfig -v

BLAS

sudo apt-get install libatlas-base-dev

1	sudo apt-get install libatlas-base-dev

其他依赖项

sudo apt-get install libgflags-dev libgoogle-glog-dev liblmdb-dev

1	sudo apt-get install libgflags-dev libgoogle-glog-dev liblmdb-dev

OpenCV

#[compiler] sudo apt-get install build-essential #[required] sudo apt-get install cmake git libgtk2.0-dev pkg-config libavcodec-dev libavformat-dev libswscale-dev #[optional] sudo apt-get install python-dev python-numpy libtbb2 libtbb-dev libjpeg-dev libpng-dev libtiff-dev libjasper-dev libdc1394-22-dev

1 2 3 4 5 6

#[compiler] sudo apt-get install build-essential   #[required] sudo apt-get install cmake git libgtk2.0-dev pkg-config libavcodec-dev libavformat-dev libswscale-dev   #[optional] sudo apt-get install python-dev python-numpy libtbb2 libtbb-dev libjpeg-dev libpng-dev libtiff-dev libjasper-dev libdc1394-22-dev

可以下载opencv的包解压，也可以用最新代码：

git clone https://github.com/Itseez/opencv.git cd ~/opencv mkdir build cd build

1 2 3 4

git clone https://github.com/Itseez/opencv.git   cd ~/opencv   mkdir build   cd build

这里可以用下载的ippicvlinux20141027.tgz放进~/opencv/3rdparty/ippicv /downloads/linux-8b449a536a2157bcad08a2b9f266828b/ (cmake之前没这个文件夹，camke的时候会执行下载，20+mb，网速快就不用管了）

cmake -D CMAKE_BUILD_TYPE=Release -D CMAKE_INSTALL_PREFIX=/usr/local #make -j 后面的数字是并行个数，cpu厉害就设大点。一般是用4 make -j7 sudo make install

1 2 3 4

cmake -D CMAKE_BUILD_TYPE=Release -D CMAKE_INSTALL_PREFIX=/usr/local   #make -j 后面的数字是并行个数，cpu厉害就设大点。一般是用4 make -j7   sudo make install

CuDNN

解压cudnn的包（官网得申请，用网盘搜索能找到最新的），有include和lib64，里面文件复制到对应/usr/local/cuda/对应文件夹里

#进到对应文件夹 sudo cp cudnn.h /usr/local/cuda/include/ #进到对应文件夹 sudo cp lib* /usr/local/cuda/lib64/ #可能要再进行一次 sudo ldconfig -v

1 2 3 4 5

#进到对应文件夹 sudo cp cudnn.h /usr/local/cuda/include/   #进到对应文件夹 sudo cp lib* /usr/local/cuda/lib64/   #可能要再进行一次 sudo ldconfig -v

不知道这里会不会有文件权限问题，暴力搞一下（这条可先不用）

sudo chmod 777 -R /usr/local/cuda/lib64

1	sudo chmod 777 -R /usr/local/cuda/lib64

Caffe

Shell

git clone https://github.com/BVLC/caffe.git cp Makefile.config.example Makefile.config

1 2	git clone https://github.com/BVLC/caffe.git   cp Makefile.config.example Makefile.config

修改Makefile.config，去掉cudnn的注释，其他的在当前应用场景可不变。

make all make test make runtest

1 2 3

make all   make test   make runtest

OK了。