linux版本安装

安装环境 ubuntu14.04LTS(官方使用版本)
环境准备

安装 g++

apt-get install g++

安装git

apt-get install git

安装ACML

官网的安装版本为:
下载地址：http://developer.amd.com/tools-and-sdks/archive/amd-core-math-library-acml/acml-downloads-resources/#download
acml-5-3-1-ifort-64bit.tgz
相关命令

tar -xzvf ./acml-5-3-1-ifort-64bit.tgz
sudo ./install-acml-5-3-1-ifort-64bit.sh

配置环境变量：

export ACML_FMA=0
export LD_LIBRARY_PATH=/opt/acml5.3.1/ifort64/lib:/opt/acml5.3.1/ifort64_mp/lib:$LD_LIBRARY_PATH

安装open mpi

相关命令

wget https://www.open-mpi.org/software/ompi/v1.10/downloads/openmpi-1.10.1.tar.gz 
tar -xzvf ./openmpi-1.10.2.tar.gz
cd openmpi-1.10.2
./configure --prefix=/usr/local/mpi
make -j all
sudo make install

配置环境变量

export PATH=/usr/local/mpi/bin:$PATH
export LD_LIBRARY_PATH=/usr/local/mpi/lib:$LD_LIBRARY_PATH

安装显卡驱动及CUDA7.5

首先停止xwindow

sudo stop lightdm

然后停止nouveau kernel driver
（ubuntu14.04 参考 http://askubuntu.com/a/451248）
修改 /etc/modprobe.d/blacklist-nouveau.conf（文件不存在的话创建它）
增加以下语句：

blacklist nouveau
blacklist lbm-nouveau
options nouveau modeset=0
alias nouveau off
alias lbm-nouveau off

重启机器后还要禁用下xwindow
然后安装最新的显卡驱动

NVIDIA-Linux-x86_64-361.42.run

安装7.5的cuda驱动
注意会提示是否安装最新的显卡驱动，如果已经安装了就选择否

cuda_7.5.18_linux.run

配置环境变量

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

验证安装是否成功

在cuda的sample目录下进行测试
进入sample目录

cd ~/NVIDIA_CUDA-7.5_Samples/

并make
成功后调用如下命令

~/NVIDIA_CUDA-7.5_Samples/1_Utilities/deviceQuery/deviceQuery

如果测试成功应该得到如下类似的显示：

/home/alexey/NVIDIA_CUDA-7.5_Samples/1_Utilities/deviceQuery/deviceQuery Starting...

 CUDA Device Query (Runtime API) version (CUDART static linking)

Detected 1 CUDA Capable device(s)

Device 0: "Quadro 600"
  CUDA Driver Version / Runtime Version          8.0 / 7.5
  CUDA Capability Major/Minor version number:    2.1
  Total amount of global memory:                 1016 MBytes (1065734144 bytes)
  ( 2) Multiprocessors, ( 48) CUDA Cores/MP:     96 CUDA Cores
  GPU Max Clock rate:                            1280 MHz (1.28 GHz)
  Memory Clock rate:                             800 Mhz
  Memory Bus Width:                              128-bit
  L2 Cache Size:                                 131072 bytes
  Maximum Texture Dimension Size (x,y,z)         1D=(65536), 2D=(65536, 65535), 3D=(2048, 2048, 2048)
  Maximum Layered 1D Texture Size, (num) layers  1D=(16384), 2048 layers
  Maximum Layered 2D Texture Size, (num) layers  2D=(16384, 16384), 2048 layers
  Total amount of constant memory:               65536 bytes
  Total amount of shared memory per block:       49152 bytes
  Total number of registers available per block: 32768
  Warp size:                                     32
  Maximum number of threads per multiprocessor:  1536
  Maximum number of threads per block:           1024
  Max dimension size of a thread block (x,y,z): (1024, 1024, 64)
  Max dimension size of a grid size    (x,y,z): (65535, 65535, 65535)
  Maximum memory pitch:                          2147483647 bytes
  Texture alignment:                             512 bytes
  Concurrent copy and kernel execution:          Yes with 1 copy engine(s)
  Run time limit on kernels:                     No
  Integrated GPU sharing Host Memory:            No
  Support host page-locked memory mapping:       Yes
  Alignment requirement for Surfaces:            Yes
  Device has ECC support:                        Disabled
  Device supports Unified Addressing (UVA):      Yes
  Device PCI Domain ID / Bus ID / location ID:   0 / 2 / 0
  Compute Mode:
     < Default (multiple host threads can use ::cudaSetDevice() with device simultaneously) >

deviceQuery, CUDA Driver = CUDART, CUDA Driver Version = 8.0, CUDA Runtime Version = 7.5, NumDevs = 1, Device0 = Quadro 600
Result = PASS

安装最新版的GPU Deployment kit

sudo chmod +x ./gdk_linux_amd64_352_79_release.run
sudo ./gdk_linux_amd64_352_79_release.run

接受默认设置即可

安装CUB

wget https://github.com/NVlabs/cub/archive/1.4.1.zip
unzip ./1.4.1.zip
sudo cp -r cub-1.4.1 /usr/local

安装cuDNN

wget http://developer.download.nvidia.com/compute/redist/cudnn/v4/cudnn-7.0-linux-x64-v4.0-prod.tgz
tar -xzvf ./cudnn-7.0-linux-x64-v4.0-prod.tgz
sudo mkdir /usr/local/cudnn-4.0
sudo cp -r cuda /usr/local/cudnn-4.0

配置环境变量

export LD_LIBRARY_PATH=/usr/local/cudnn-4.0/cuda/lib64:$LD_LIBRARY_PATH

可选 安装open cv

需要的空间及时间比较大，暂时不安装，可参看github中的官方说明

安装zib及libzib

apt-get install zlib1g-dev

wget http://nih.at/libzip/libzip-1.1.2.tar.gz
tar -xzvf ./libzip-1.1.2.tar.gz
cd libzip-1.1.2
./configure
make -j all
sudo make install

配置环境变量

export LD_LIBRARY_PATH=/usr/local/lib:$LD_LIBRARY_PATH

由源码安装cntk

不包括1bit-sgd code

git clone https://github.com/Microsoft/cntk

包括1bit-sgd code

git clone --recursive https://github.com/Microsoft/cntk/

编译代码

cd ~/Repos/cntk
mkdir build/release -p
cd build/release
../../configure --1bitsgd=yes

不加–1bitsgd=yes 则编译没有1bitsgd的版本

debug 版本的编译方法：
../../configure –with-buildtype=debug

没有报错的情况下运行

make -j all

export PATH=$HOME/Repos/cntk/build/release/bin:$PATH

验证ok

cntk configFile=../Config/Simple.cntk

验证cntk with gpu

cntk configFile=../Config/Simple.cntk deviceId=auto &> out

cat out | grep Builder
Expected output in this case is:

    SimpleNetworkBuilder = [
    SimpleNetworkBuilder = [
SimpleNetworkBuilder Using GPU 0

运行滨州书库的例子

cntk configFile=Config/rnn.cntk currentDirectory=Data deviceId=auto