國內(nèi)的thinkpad t系列一般會配一塊N卡田轧，想著雖然卡不咋地，但是做一些簡單的實驗還是可以的，于是決定在這臺本子上面裝CUDA和caffe來練習一下。下面是步驟伊履，前方有坑，要仔細讀：

• 安裝NVIDIA的驅(qū)動款违，使用ubuntu自己的driver安裝：
  settings -> software & updates -> Additional Drivers -> 選擇安裝NVIDIA驅(qū)動
  雖然有手動安裝的方法唐瀑，但是我覺得這樣最保險，一定不會裝錯插爹。

• 去NVIDIA官網(wǎng)下載CUDA8.0哄辣，注意用.run文件，而不是.deb赠尾，據(jù)說這樣坑比較少一些力穗。。气嫁。

• log out出去当窗，然后ctrl + alt + f1進入命令行模式：
  首先，禁止X server桌面寸宵，不然無法繼續(xù)安裝下去崖面。

sudo service lightdm stop

然后，按照英偉達官網(wǎng)的指令：

sudo sh cuda.***.run (看你下的哪個版本)

我第一次裝的時候沒有指明tmpdir也裝成了梯影，但是因為下面有一步出錯了卸載了重新裝巫员，沒有加tmpdir的話會提示 disk space check has failed. Installation cannot continue. 這種情況就要自己指定一個解壓的地方來存放臨時安裝文件了。后面可以寫 --tmpdir=...(根據(jù)自己的情況)不過我后來很囧的發(fā)現(xiàn)我出現(xiàn)這個問題是因為后面自己在/目錄下面拷了一個有點大的數(shù)據(jù)集甲棍。简识。。rm之后就好了，不用指明tmpdir七扰，就直接在/下就可以奢赂。<br />
接下來，安裝的時候很重要的一點是別盲目選y!!!第一個選項問你要不要裝一個Driver戳寸，要選no呈驶，要選no，要選no疫鹊，重要的話說三遍袖瞻。因為這個就是第一步干的事，不能用它這里的driver拆吆，不然很可能不匹配聋迎，reboot之后就會出現(xiàn)很蛋疼的問題，什么low graphics blabla的...桌面都出不來了枣耀，只能卸掉nvidia的驅(qū)動重新來霉晕。

• 安裝完成后reboot，檢驗一下是不是裝好了捞奕。

lspci | grep -i nvidia

如果能檢測到的話說明驅(qū)動沒問題牺堰。再看一下現(xiàn)在的主顯卡：

prime-select query

test CUDA:

nvidia-smi

如果可以顯示GPU當前信息的話，那么CUDA就裝好了颅围。
還可以把sample都編譯一下伟葫，進入/home/username/NVIDIA...

sudo make

一堆warning...哎...<br >這里要設置一下環(huán)境變量，

sudo gedit /etc/profile

文件末尾添加PATH=/usr/local/cuda/bin:$PATHexport PATH <br >保存完成后院促，執(zhí)行如下命令使環(huán)境變量立即生效：

source /etc/profile

然后還需要添加lib的路徑：

sudo gedit /etc/ld.so.conf.d/cuda.conf

在文件中寫入如下內(nèi)容然后保存：
/usr/local/cuda/lib64<br >之后執(zhí)行如下命令使之生效：

sudo ldconfig

ldconfig命令的用途筏养，主要是在默認搜尋目錄(/lib和/usr/lib)以及動態(tài)庫配置文件/etc/ld.so.conf內(nèi)所列的目錄下，搜索出可共享的動態(tài)鏈接庫(格式如前介紹,lib.so)常拓，進而創(chuàng)建出動態(tài)裝入程序(ld.so)所需的連接和緩存文件渐溶。緩存文件默認為/etc/ld.so.cache，此文件保存已排好序的動態(tài)鏈接庫名字列表弄抬。我們之前修改過cuda.conf茎辐，為了讓鏈接可以找到這里來，才需要這個ldconfig的命令掂恕。
或者這樣也可以：

echo 'export PATH=/usr/local/cuda/bin:$PATH' >> ~/.bashrc
echo 'export LD_LIBRARY_PATH=/usr/local/cuda/lib64:$LD_LIBRARY_PATH' >> ~/.bashrcsource ~/.bashrc

• 再把cuDNN裝了：
  依然取去官網(wǎng)下荔茬，我下的是CuDNN Library for Linux v5.1，下載需要注冊一下...CuDNN的作用是對CUDA進行一些修改和優(yōu)化竹海，使得更適合于對神經(jīng)網(wǎng)絡進行計算。
  <br />下載后解壓丐黄，然后進入該目錄斋配，把所有的lib復制到/usr/local/cuda/lib64/下，把頭文件 cudnn.h復制到 /usr/local/cuda/include/下。

sudo cp lib* /usr/local/cuda/lib64/
sudo cp cudnn.h /usr/local/cuda/include/

然后更新軟連接：

cd /usr/local/cuda/lib64/
sudo rm -rf libcudnn.so libcudnn.so.5
sudo ln -s libcudnn.so.5.1.5 libcudnn.so.5
sudo ln -s libcudnn.so.5 libcudnn.so

我安裝的是5.1.5的艰争，大家要看下自己cudnn解壓出來的lib下面的版本號坏瞄，不可以盲目復制...

• 安裝一些依賴項

sudo apt-get install freeglut3-dev libx11-dev libxmu-dev libxi-dev libglu1-mesa-dev
sudo apt-get install libgtk2.0-dev pkg-config libavcodec-dev libavformat-dev libswscale-dev
sudo apt-get install python-dev python-numpy libtbb2 libtbb-dev libjpeg-dev libpng-dev libtiff-dev libjasper-dev libdc1394-22-dev
sudo apt-get install libprotobuf-dev libleveldb-dev libsnappy-dev libopencv-dev libboost-all-dev libhdf5-serial-dev libgflags-dev libgoogle-glog-dev liblmdb-dev protobuf-compiler

• 安裝python的必備

sudo apt-get install python-dev python-pip
cd python
for req in $(cat requirements.txt); do sudo pip install $req; done

我的pip一直出些問題，后來我就用sudo easy-install pip安裝了pip甩卓。這個apt-get安裝出來的很蛋疼不知道為什么...

• 安裝caffe

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

要安裝BLAS作為caffe的matrix運算支持鸠匀。可以選擇安裝Atlas逾柿。

sudo apt-get install libatlas-base-dev
echo 'export LD_LIBRARY_PATH=/usr/local/lib:$LD_LIBRARY_PATH' >> ~/.bashrc

編譯之前修改Makefile.config：
由于需要CuDNN缀棍，所以把 # USE_CUDNN := 1的注釋去掉。還有Anaconda2的路徑修改等机错。我根據(jù)自己的安裝修改后的內(nèi)容如下：

## 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
    # MATLAB_DIR := /matlab

    # 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 \

    # 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 

    # 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

    # 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 ?= @

caffe目錄下：

make all -j16
make test
make runtest

在進行make runtest的時候可能會報錯說找不到libhdf5.so.10的錯誤爬范。我解決這個的方式是去/usr/lib/下把libhdf5.so.7和libhdf5_hl.so.7該成了.10。弱匪。青瀑。。因為我發(fā)現(xiàn)我即使把那些依賴都裝好了萧诫，還是.7斥难，如果有更好的解決方案請告訴我，謝謝～<br />根據(jù)需求編譯matcaffe和pycaffe

make matcaffe
make pycaffe

matcaffe要能找到mex文件帘饶，所以先把matlab裝好然后把matlab的路徑填到Makefile.config里面去哑诊。還有在make pycaffe的時候，因為我是用的anaconda2尖奔，Python.h就在anaconda2/include/python2.7/下面搭儒，所以這個路徑要包含進去，在PYTHON_INCLUDE下面ANACONDA的第二行取消注釋提茁。

• 運行mnist demo

sh data/mnist/get_mnist.sh
sh examples/mnist/create_mnist.sh
sh examples/mnist/train_lenet.sh