本文旨在Ubuntu16.04下的caffe環境搭建。
顯卡：TESLA K80
一、系統安裝
這里與大多系統安裝步驟無異，因而略過。大抵的一些注意事項是
1.時區選擇的時候需要斷網；
2.大陸地區的需要換源
3.apt-get update
選擇的鏡像：Ubuntu16.04-desktop
二、驅動安裝
之前安裝Ubuntu14.04下的驅動時，都是直接用CUDA安裝包里自帶的顯卡驅動。但在當前目標配置下，安裝驅動的時候會報couldn't locate the kernel files之類的錯誤。反復無法解決。遂選擇手動安裝顯卡驅動。
在NVIDIA官網下載對應的驅動版本。這里選擇用runfile安裝。
1.禁用nouveau
vi /etc/modprobe.d/blacklist-nouveau.conf
加入文本
blacklist nouveau
options nouveau modeset=0
更新內核
update-initramfs -u
重啟之后，查看nouveau是否禁用成功
lsmod | grep nouveau
2.禁用x server
如是lightdm，則如下命令，其他以此類推
service lightdm stop
如果該命令之后x server沒有成功關閉，可嘗試
pkill x
3.安裝驅動
注意不要安裝opengl libs（否則桌面環境不正常）
chmod a+x NVIDIA.run
./NVIDIA.run —no-opengl-files
調用以下命令，若顯示顯卡信息，則安裝驅動成功。
nvidia-smi
三、CUDA8.0
CUDA歷史版本，這里選擇CUDA Toolkit 8.0 GA2。
同樣別安裝opengl libs，其余選項皆為默認或yes
chmod a+x cuda.run
./cuda.run —no-opengl-libs
設置環境變量和動態鏈接庫
vi? ?/etc/profile
在文件末尾加入：
export PATH = /usr/local/cuda/bin:$PATH
再創建鏈接文件
vi /etc/ld.so.conf.d/cuda.conf
加入文本
/usr/local/cuda/lib64
再執行以下命令使鏈接生效
ldconfig
編譯測試程序，若出現顯卡信息，則安裝成功。
cd /usr/local/cuda-8.0/samples/1_Utilities/deviceQuery?
make
?./deviceQuery
四、CUDNN6.0
CUDNN下載地址，這里選擇6.0版本。
tar zxvf cudnn-8.0-linux-x64-v6.0.tgz
cd cuda
cp include/cudnn.h /usr/local/include
cp lib64/libcudnn.* /usr/local/lib
ln -sf /usr/local/lib/libcudnn.so.6.0.21 /usr/local/lib/libcudnn.so.6
ln -sf /usr/local/lib/libcudnn.so.6 /usr/local/lib/libcudnn.so
ldconfig -v
若出現gcc版本過高，而編譯caffe-master報錯的話，
vi /usr/local/cuda/include/host_config.h
搜索： #error -- unsupported GNU version! gcc versions later than 5.3 are not supported!
修改為： //#error -- unsupported GNU version! gcc versions later than 5.3 are not supported!
五、Matlab2016b
關于 GCC 和 G++ 版本問題
Matlab 2014a gcc/g++ 4.7.x, Matlab 2016a gcc/g++ 4.9.x
Ubuntu 15.04 gcc/g++ 4.9.x, Ubuntu 16.04 gcc/g++ 5.4.x
原則上Matlab需要和Ubuntu版本一致，由于CUDA 8只支持16.04,而且需要GCC 5.4.x 進行編譯，而CUDA 7.5不支持 Ubuntu 16.04 因此Matlab會有一些奇葩，有時按照降級（或強制安裝）的方法可以正常使用，有時卻會報錯，懷疑和顯卡驅動有關。
引自宇宙騎士歐老師
因而選擇了2016b版本。
將鏡像文件掛載到image monitor上。將所有文件拷貝到Home/Matlab。
chmod a+x Matlab -R
./install
相關選項：
-?不使用Internet安裝
-?密鑰：09806-07443-53955-64350-21751-41297
-?安裝完成后，運行Matlab安裝目錄下bin/glx64/activate_matlab.sh激活
- Crack文件夾下license_standalone.lic為許可證文件
-?復制MATLAB Production Server\R2016b\bin\glx64下的文件到對應目錄下，并替換源文件
-?激活完畢，可運行matlab
六、BLAS
安裝MKL，以學生身份下載Student版，填好各種信息，可以直接下載，同時會給你一個郵件告知序列號。下載完之后，要把文件解壓到home文件夾(注意任何一級文件夾不能包含空格，否則安裝會失敗)。這里下載的是2017版本。
tar zxvf parallel_studio_xe_2017.tar.gz?
chmod a+x parallel_studio_xe_2017-R
sh install_GUI.sh
環境配置，新建conf文件
vi /etc/ld.so.conf.d/intel_mkl.conf
并鍵入
/opt/intel/lib/intel64
/opt/intel/mkl/lib/intel64
七、OpenCV3.1
安裝依賴項
apt-get install build-essentialsudo apt-get install cmake git libgtk2.0-dev pkg-config libavcodec-dev libavformat-dev libswscale-dev
apt-get install python-dev python-numpy libtbb2 libtbb-dev libjpeg-dev libpng-dev libtiff-dev libjasper-dev libdc1394-22-devsudo apt-get install –assume-yes libopencv-dev libdc1394-22 libdc1394-22-dev libjpeg-dev libpng12-dev libtiff5-dev libjasper-dev libavcodec-dev libavformat-dev libswscale-dev libxine2-dev libgstreamer0.10-dev libgstreamer-plugins-base0.10-dev libv4l-dev libtbb-dev libqt4-dev libfaac-dev libmp3lame-dev libopencore-amrnb-dev libopencore-amrwb-dev libtheora-dev libvorbis-dev libxvidcore-dev x264 v4l-utils unzip
apt-get install ffmpeg libopencv-dev libgtk-3-dev python-numpy python3-numpy libdc1394-22 libdc1394-22-dev libjpeg-dev libpng12-dev libtiff5-dev libjasper-dev libavcodec-dev libavformat-dev libswscale-dev libxine2-dev libgstreamer1.0-dev libgstreamer-plugins-base1.0-dev libv4l-dev libtbb-dev qtbase5-dev libfaac-dev libmp3lame-dev libopencore-amrnb-dev
下載OpneCV源文件
wget -O opencv-3.1.0.zip http://sourceforge.net/projects/opencvlibrary/files/opencv-unix/3.1.0/opencv-3.1.0.zip/download
或者從Github clone
mkdir opencv
cd opencv ?
git clone?https://github.com/opencv/opencv.git?
git clone?https://github.com/opencv/opencv_contrib.git
編譯和安裝
mkdir build
cd build
cmake -D CMAKE_BUILD_TYPE=RELEASE -D CMAKE_INSTALL_PREFIX=/usr/local -D WITH_TBB=ON -D BUILD_NEW_PYTHON_SUPPORT=ON -D WITH_V4L=ON -D INSTALL_C_EXAMPLES=ON -D INSTALL_PYTHON_EXAMPLES=ON -D BUILD_EXAMPLES=ON -D WITH_QT=ON -D WITH_OPENGL=ON ..
make -j4
sudo make install
sudo sh -c 'echo "/usr/local/lib" > /etc/ld.so.conf.d/opencv.conf'
sudo ldconfig
這里可能遇到的問題包括：
- ippicv無法下載
- cuda8.0不支持
解決方案：
- 自行下載ippicv_linux_20151201，將文件拷貝到opencv-3.1.0/3rdparty/ippicv/downloads/linux-808b791a6eac9ed78d32a7666804320e/路徑下
-?modules/gpu/src/graphcuts.cpp
#if !defined (HAVE_CUDA) || defined (CUDA_DISABLER)
改成
#if !defined (HAVE_CUDA) || defined (CUDA_DISABLER) || (CUDART_VERSION >= 8000)
八、其他依賴項
1. Google Logging Library（glog）下載地址
tar zxvf glog-0.3.3.tar.gz
./configure
make
sudo make install
如果沒有權限就
chmod a+x glog-0.3.3 -R
2.其他依賴項
sudo apt-get install -y libprotobuf-dev libleveldb-dev libsnappy-dev libopencv-dev libboost-all-dev libhdf5-serial-dev libgflags-dev libgoogle-glog-dev liblmdb-dev protobuf-compiler protobuf-c-compiler protobuf-compiler python-numpy python-scipy python-matplotlib python-sklearn python-skimage python-h5py python-protobuf python-leveldb python-networkx python-nose python-pandas python-gflags cython ipython
九、Caffe
下載源文件，進行配置
cp Makefile.config.example Makefile.config
vi? Makefile.config
1.啟用CUDNN
USE_CUDNN := 1
2.配置引用文件（Ubuntu16.04下，文件位置發生變化）
INCLUDE_DIRS := $(PYTHON_INCLUDE) /usr/local/include /usr/lib/x86_64-linux-gnu/hdf5/serial/include
LIBRARY_DIRS := $(PYTHON_LIB) /usr/local/lib /usr/lib /usr/lib/x86_64-linux-gnu/hdf5/serial
3.啟用Intel Parallel Studio XE 2017
BLAS := mkl
4.配置路徑，實現caffe對Python和Matlab接口的支持
PYTHON_LIB := /usr/local/lib
MATLAB_DIR := /usr/local/MATLAB/R2016b
5.啟用Opencv
OPENCV_VERSION =3
這里列出一份配置參考
## 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 through *_61 lines for compatibility.
# For CUDA < 8.0, comment the *_60 and *_61 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_52,code=sm_52 \
-gencode arch=compute_60,code=sm_60 \
-gencode arch=compute_61,code=sm_61 \
-gencode arch=compute_61,code=compute_61
# BLAS choice:
# atlas for ATLAS (default)
# mkl for MKL
# open for OpenBlas
BLAS := mkl
# 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/R2016b
# 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)/anaconda
# PYTHON_INCLUDE := $(ANACONDA_HOME)/include \
# $(ANACONDA_HOME)/include/python2.7 \ # $(ANACONDA_HOME)/lib/python2.7/site-packages/numpy/core/include
# Uncomment to use Python 3 (default is Python 2)
# PYTHON_LIBRARIES := boost_python3 python3.5m
# PYTHON_INCLUDE := /usr/include/python3.5m \
# /usr/lib/python3.5/dist-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
INCLUDE_DIRS := $(PYTHON_INCLUDE) /usr/local/include /usr/lib/x86_64-linux-gnu/hdf5/serial/include
LIBRARY_DIRS := $(PYTHON_LIB) /usr/local/lib /usr/lib /usr/lib/x86_64-linux-gnu/hdf5/serial
# 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
# NCCL acceleration switch (uncomment to build with NCCL)
# https://github.com/NVIDIA/nccl (last tested version: v1.2.3-1+cuda8.0)
# USE_NCCL := 1
# 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
# N.B. both build and distribute dirs are cleared on `make clean` 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 ?= @
編譯caffe-master
make all -j16
make test -j16
make runtest -j16
make pycaffe -j16
make matcaffe -j16
十、測試
這里用mnist數據集進行測試。如若下載數據集不方便，附上云盤下載 。
不做具體說明。
sh data/mnist/get_mnist.sh
sh examples/mnist/create_mnist.sh
sh examples/mnist/train_lenet.sh
至此，配置完畢。
