Ubuntu14.04+cuda7.5+caffe+OpenCV2.4.9+cudnn7.5+Anaconda2安装配置

版权声明：本文为博主原创文章，未经博主允许不得转载。 https://blog.csdn.net/abc8730866/article/details/70142113

系统出问题不得不重装系统，所有caffe相关也都需重装，在此做个记录。
环境：Ubuntu14.04/台式机集成显卡+独立显卡gtx960
准备好的文件：

1.安装cuda7.5

a）禁用nouveau

$ cd /etc/modprobe.d/
$ sudo geidit blacklist.conf

在/etc/modprobe.d/blacklist.conf最后一行加上：

blacklist nouveau
options nouveau modeset=0

b)关闭GUI：

$ sudo service lightdm stop

c) Alt + ctrl +F1进入命令行
d) 更改CUDA Toolkit执行权限：
切换到cuda安装文件的路径

$ sudo chmod +x cuda_7.5.18_linux.run

e) 执行CUDA安装文件

$ sudo ./cuda_7.5.18_linux.run --no-opengl-libs

f) 切换回GUI：

$ sudo service lightdm start

g) 成功切回GUI后,添加环境变量PATH：

$ sudo gedit /etc/profile

h) 在/etc/profile文件最后添加

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

i) 然后source一下，使之生效

$ source /etc/profile   

j) 检查上述的环境变量是否设置成功。
终端中输入 :
$ env
在输出的环境变量中检查有无上述 h) 中设置的变量，如果有则代表设置成功。
k) 再然后就是添加共享库变量:

$ cd /etc/ld.so.conf.d/
$ touch cuda.conf
$ sudo gedit cuda.conf

在 /etc/ld.so.conf.d/加入文件 cuda.conf, 内容如下

/usr/local/cuda-7.5/lib64

执行下列命令使之立刻生效

$ sudo ldconfig

---

2.检验cuda是否安装成功

a) 检查 NVIDIA Driver是否安装成功
终端输入 ：

$ cat /proc/driver/nvidia/version 

会输出NVIDIA Driver的版本号
b) 检查 CUDA Toolkit是否安装成功
终端输入 ：

$ nvcc –V 

会输出CUDA的版本信息
c) 尝试编译cuda提供的例子
切换到例子存放的路径，默认路径是 ~/NVIDIA_CUDA-7.5_Samples
（即 /home/XX/ NVIDIA_CUDA-7.5_Samples）

$ cd ~/NVIDIA_CUDA-7.5_Samples/

然后终端输入：

$ make

如果出现错误的话，则会立即报错停止，否则会开始进入编译阶段。
我的第一次运行时出现了报错，提示的错误信息是系统中没有g++
然后在终端运行 :
$ sudo apt-get install g++
安装完g++后，再make就正常了 。整个编译的时间持续比较长，耐心等待。
d) 运行编译生成的二进制文件。
编译后的二进制文件 默认存放在~/NVIDIA_CUDA-7.5_Samples/bin/x86_64/linux/release中。
切换路径 ：

$ cd ~/NVIDIA_CUDA-7.5_Samples/bin/x86_64/linux/release

终端输入 ：
$ ./deviceQuery
若正确显示安装的cuda7.5即GPU信息，则表示安装成功。

---

3.安装依赖库

包含了boost opencv leveldb protobuf 等:

$ sudo apt-get install libprotobuf-dev libleveldb-dev libsnappy-dev libopencv-dev libboost-all-dev libhdf5-serial-dev

$ sudo apt-get install libgflags-dev libgoogle-glog-dev liblmdb-dev protobuf-compiler

BLAS安装：

$ sudo apt-get install libatlas-base-dev

---

4.Caffe安装

a) 安装git

$ sudo add-apt-repository ppa:git-core/ppa

中间暂停时，按回车键Enter继续安装。

$ sudo apt-get update

$ sudo apt-get install git     

安装下载完成后，可以使用下面的命令行，确认git的版本：

$ git --version 

b) 下载caffe（注意下载的位置，下载到/home/just）：

$ git clone https://github.com/BVLC/caffe.git

Cd进home/just/caffe，然后编译caffe：

$ cd ~/caffe
$ sudo cp Makefile.config.example Makefile.config
$ sudo make all
$ sudo make test  
$ sudo make runtest  

---

5.安装OpenCV

a) 下载OpenCV
b) 解压到任意目录

$ unzip opencv-2.4.9.zip

c) 进入源码目录，创建release目录

$ cd opencv-2.4.9

$ mkdir release  

d) 可以看到在OpenCV目录下，有个CMakeLists.txt文件，需要事先安装一些软件

$ build-essential cmake libgtk2.0-dev pkg-config python-dev python-numpy libavcodec-dev libavformat-dev libswscale-dev  

e) 进入release目录，安装OpenCV是所有的文件都会被放到这个release目录下

$ cd release  

f) cmake编译OpenCV源码，安装所有的lib文件都会被安装到/usr/local目录下

$ sudo apt-get  libqt4-dev 

$ cmake -D CMAKE_BUILD_TYPE=RELEASE -D CMAKE_INSTALL_PREFIX=/usr/local -D WITH_CUDA=OFF -D WITH_OPENMP=ON -D WITH_QT=ON ..

注:这里-D的作用是设置编译的选项，如选择release模式， 输出的目录，不编译cuda模块，编译openMP与QT模块，这里的QT模式带有很强的可视化效果，推荐。但需要在第一步中加上 libqt4-dev 安装QT环境
g) 安装

$ sudo make install    

---

6.测试 OpenCV

(1) Create a directory DisplayImagefor test project:

$ mkdir DisplayImage  
$ cd DisplayImage  

(2) Create DisplayImage.cpp and edit it: gedit DisplayImage.cpp
Then edit the DisplayImage.cpp:

#include <stdio.h>
#include <opencv2/opencv.hpp>
using namespace cv;
int main(int argc, char** argv )
{
  if ( argc != 2 )
  {
    printf("usage: DisplayImage.out <Image_Path>\n");
    return -1;
  }
  Mat image;
  image = imread( argv[1], 1 );
  if ( !image.data )
  {
    printf("No image data \n");
    return -1;
  }
  namedWindow("Display Image", WINDOW_AUTOSIZE );
  imshow("Display Image", image);
  waitKey(0);
  return 0;
}

(3) Create a CMake file:

$ gedit CMakeLists.txt  

Then edit the CMakeLists.txt:

cmake_minimum_required(VERSION 2.8)  
project(DisplayImage)  
find_package(OpenCV REQUIRED)  
add_executable(DisplayImage DisplayImage.cpp)  
target_link_libraries(DisplayImage ${OpenCV_LIBS})  

(4) Generate the executable file:

$ cmake .  

$ make  

(5) Execute it:

$ ./DisplayImage lena.jpg  

成功则显示图像。

---

7.cudnn

6.cudnn
1.进官网下载文件（cudnn-7.5-linux-x64-v5.0-ga.tgz）
在终端中切换到文件所在文件夹，输入下面指令：

$ sudo tar xvf cudnn-7.5-linux-x64-v5.0-ga.tgz
$ cd cuda/include
$ sudo cp *.h /usr/local/include/
$ cd ../lib64
$ sudo cp lib* /usr/local/lib/
$ cd /usr/local/lib
$ sudo chmod +r libcudnn.so.5.0.5 
$ /usr/local/lib$ sudo ln -sf libcudnn.so.5.0.5 libcudnn.so.5
$ sudo ln -sf libcudnn.so.5 libcudnn.so^C
$ sudo ldconfig    

然后切换到caffe根目录下，将Makefile.config中的USE_CUDNN行前的#去掉：

$ cd ~/caffe
$ sudo gedit Makefile.config

将Makefile.config中的USE_CUDNN行前的#去掉。
保存后重新编译：

$ cd ~/caffe
$ sudo make clean
$ sudo make all
$ sudo make test  
$ sudo make runtest 

遇到过error：

.build_release/tools/caffe: error while loading shared libraries: libcudart.so.7.0: cannot open shared object file: No such file or directory

解决办法

sudo ldconfig /usr/local/cuda-7.5/lib64

---

8.配置pycaffe

1. 安装anaconda2
   去官网下载，下载2.7版本（Anaconda2-4.2.0-Linux-x86_64.sh）。

bash Anaconda2-4.2.0-Linux-x86_64.sh    

安装的时候，最后一定要选yes。
安装完毕打开home/just/.bashrc发现，它自动在最后添加了：
#Add by anaconda2 4.2.0 installer
Export PATH=”/home/just/anaconda2/bin:$PATH”
修改Makefile.config:

cd ~/caffe
sudo gedit Makefile.config

修改过的如下：

## Refer to http://caffe.berkeleyvision.org/installation.html
# Contributions simplifying and improving our build system are welcome!

# cuDNN acceleration switch (uncomment to build with cuDNN).
USE_CUDNN := 1

# CPU-only switch (uncomment to build without GPU support).
# CPU_ONLY := 1

# uncomment to disable IO dependencies and corresponding data layers
# USE_OPENCV := 0
# USE_LEVELDB := 0
# USE_LMDB := 0

# uncomment to allow MDB_NOLOCK when reading LMDB files (only if necessary)
#    You should not set this flag if you will be reading LMDBs with any
#    possibility of simultaneous read and write
# ALLOW_LMDB_NOLOCK := 1

# Uncomment if you're using OpenCV 3
# OPENCV_VERSION := 3

# To customize your choice of compiler, uncomment and set the following.
# N.B. the default for Linux is g++ and the default for OSX is clang++
# CUSTOM_CXX := g++

# CUDA directory contains bin/ and lib/ directories that we need.
CUDA_DIR := /usr/local/cuda
# On Ubuntu 14.04, if cuda tools are installed via
# "sudo apt-get install nvidia-cuda-toolkit" then use this instead:
# CUDA_DIR := /usr

# CUDA architecture setting: going with all of them.
# For CUDA < 6.0, comment the *_50 lines for compatibility.
CUDA_ARCH := -gencode arch=compute_20,code=sm_20 \
        -gencode arch=compute_20,code=sm_21 \
        -gencode arch=compute_30,code=sm_30 \
        -gencode arch=compute_35,code=sm_35 \
        -gencode arch=compute_50,code=sm_50 \
        -gencode arch=compute_50,code=compute_50

# BLAS choice:
# atlas for ATLAS (default)
# mkl for MKL
# open for OpenBlas
BLAS := atlas
# Custom (MKL/ATLAS/OpenBLAS) include and lib directories.
# Leave commented to accept the defaults for your choice of BLAS
# (which should work)!
# BLAS_INCLUDE := /path/to/your/blas
# BLAS_LIB := /path/to/your/blas

# Homebrew puts openblas in a directory that is not on the standard search path
# BLAS_INCLUDE := $(shell brew --prefix openblas)/include
# BLAS_LIB := $(shell brew --prefix openblas)/lib

# This is required only if you will compile the matlab interface.
# MATLAB directory should contain the mex binary in /bin.
# MATLAB_DIR := /usr/local
# MATLAB_DIR := /Applications/MATLAB_R2012b.app

# NOTE: this is required only if you will compile the python interface.
# We need to be able to find Python.h and numpy/arrayobject.h.
# PYTHON_INCLUDE := /usr/include/python2.7 \
        /usr/lib/python2.7/dist-packages/numpy/core/include
# Anaconda Python distribution is quite popular. Include path:
# Verify anaconda location, sometimes it's in root.
ANACONDA_HOME := $(HOME)/anaconda2
PYTHON_INCLUDE := $(ANACONDA_HOME)/include \
        $(ANACONDA_HOME)/include/python2.7 \
        $(ANACONDA_HOME)/lib/python2.7/site-packages/numpy/core/include \

# We need to be able to find libpythonX.X.so or .dylib.
# PYTHON_LIB := /usr/lib
PYTHON_LIB := $(ANACONDA_HOME)/lib

# Homebrew installs numpy in a non standard path (keg only)
# PYTHON_INCLUDE += $(dir $(shell python -c 'import numpy.core; print(numpy.core.__file__)'))/include
# PYTHON_LIB += $(shell brew --prefix numpy)/lib

# Uncomment to support layers written in Python (will link against Python libs)
WITH_PYTHON_LAYER := 1

# Whatever else you find you need goes here.
INCLUDE_DIRS := $(PYTHON_INCLUDE) /usr/local/include
LIBRARY_DIRS := $(PYTHON_LIB) /usr/local/lib /usr/lib

# If Homebrew is installed at a non standard location (for example your home directory) and you use it for general dependencies
# INCLUDE_DIRS += $(shell brew --prefix)/include
# LIBRARY_DIRS += $(shell brew --prefix)/lib

# Uncomment to use `pkg-config` to specify OpenCV library paths.
# (Usually not necessary -- OpenCV libraries are normally installed in one of the above $LIBRARY_DIRS.)
# USE_PKG_CONFIG := 1

BUILD_DIR := build
DISTRIBUTE_DIR := distribute

# Uncomment for debugging. Does not work on OSX due to https://github.com/BVLC/caffe/issues/171
# DEBUG := 1

# The ID of the GPU that 'make runtest' will use to run unit tests.
TEST_GPUID := 0

# enable pretty build (comment to see full commands)
Q ?= @

开始编译：

$ cd ~/caffe/
$ sudo make clean 
$ sudo make all -j8
$ sudo make test -j8
$ sudo make runtest -j8
$ make pycaffe -j8

编译runtest的时候，遇到这样的错误：
.build_release/test/test_all.testbin: error while loading shared libraries: libhdf5.so.10: cannot open shared object file: No such file or directory
这是因为 libhdf5.so的版本问题，你可以进入/usr/lib/x86_64-linux-gnu看一下，你的libhdf5.so.x中的那个x是多少，比如我的是libhdf5.so.7
因此可以执行下面几行代码解决:

# cd /usr/lib/x86_64-linux-gnu
# sudo ln -s libhdf5.so.7 libhdf5.so.10
# sudo ln -s libhdf5_hl.so.7 libhdf5_hl.so.10
# sudo ldconfig

最终查看python接口是否编译成功：
进入python环境，进行import操作

$  python
>>> import caffe

如果没有提示错误，则编译成功。
报错：no module named caffe
解决方法：将anacanda与caffe头文件进行链接。
打开.bashrc：

$ cd ~
$ sudo gedit .bashrc

添加：
export PYTHONPATH=”/home/just/caffe/python:$PYTHONPATH”
在进入python环境，import caffe,
报错：ImportError:No module named google.protobuf.internal
解决方法：

$ pip install protobuf
$ /home/just/anaconda2/bin/pip install protobuf

尽情的：

$ python   
>>Import caffe

$ Ipython 
>>Import caffe

由于安装了anaconda，jupyter notebook就已经自动安装好,可以：

$ jupyter notebook

就会在浏览器中打开notebook。

---