Ubuntu16.04 + gtx1060 + cuda8.0 + cudnn5.1 + caffe + Theano + Tensorflow

參考 ubuntu16.04+gtx1060+cuda8.0+caffe安裝、測試經歷 ，細節處有差別。

 

首先說明，這是在臺式機上的安裝測試經歷，首先安裝的win10，而後安裝ubuntu16.04雙系統，顯卡爲GTX1060 
臺式機顯示器接的是GTX1060 HDMI口，win10上首先安裝了最新的GTX1060驅動375

---

廢話很少說，上車吧，少年

1、首先安裝nvidia顯卡驅動

1. 我是1080P的顯示器，在沒有安裝顯卡驅動前，ubuntu分辨率很低，能夠手動修改一下grub文件，提升分辨率，在終端輸入

   sudo vim /etc/default/grub
   找到如下行 
   # The resolution used on graphical terminal
# note that you can use only modes which your graphic card supports via VBE
# you can see them in real GRUB with the command 'vbeinfo'
# GRUB_GFXMODE=640×480
   按a進入插入模式，增長下面一行 
   GRUB_GFXMODE=1920×1080 #這裏分辨率自行設置 
   按esc退出插入模式，按:wq保存退出 
   在終端編輯 
   sudo update-grub
   更新grub 
   從新啓動ubuntu使之生效 
   
   

2. 進入ubuntu系統設置-軟件與更新-附加驅動

   
   安裝以後重啓系統讓GTX1060顯卡驅動生效 

3. 測試

   終端輸入 
   nvidia-smi 
   顯示效果以下圖表示安裝成功 
   

 

 

---

2、cuda安裝

1. 下載cuda_8.0.61_375.26_linux.run 和 cudnn-8.0-linux-x64-v5.1.tgz

   這裏我提供了百度網盤，這兩個文件我先在win10下下載好，並用u盤拷貝到ubuntu的下載目錄下 
   
   

2. 安裝cuda8.0

   終端輸入 
   cd 下載/ 
sh cuda_8.0.27_linux.run --override 
   啓動安裝程序，一直按空格到最後，輸入accept接受條款 (或者按 Q)
   輸入n不安裝nvidia圖像驅動，以前已經安裝過了 
   輸入y安裝cuda 8.0工具 
   回車確認cuda默認安裝路徑：/usr/local/cuda-8.0 
   輸入y用sudo權限運行安裝，輸入密碼 
   輸入y或者n安裝或者不安裝指向/usr/local/cuda的符號連接 
   輸入y安裝CUDA 8.0 Samples，以便後面測試 
   回車確認CUDA 8.0 Samples默認安裝路徑：/home/yt（yt是個人用戶名），該安裝路徑測試完能夠刪除 
   
   

3. 安裝cudnn v5.1

   終端輸入 
   

   cd 下載/ 
   tar zxvf cudnn-8.0-linux-x64-v5.1.tgz

   解壓在下載目錄下產生一個cuda目錄 

   cd cuda/include/ 
   sudo cp cudnn.h /usr/local/cuda/include/   #複製頭文件 
   cd ../lib64   #打開lib64目錄 
   sudo cp lib* /usr/local/cuda/lib64/     #複製庫文件 
   sudo chmod a+r /usr/local/cuda/include/cudnn.h
   sudo chmod a+r /usr/local/cuda/lib64
   sudo chmod a+r /usr/local/cuda/lib64/libcudnn*   #給全部用戶增長這些文件的讀權限 

    

4. 創建軟連接

   終端輸入 

   cd /usr/local/cuda/lib64/ 
   sudo rm -rf libcudnn.so libcudnn.so.5 
   sudo ln -s libcudnn.so.5.1.10 libcudnn.so.5   #具體看版本
   sudo ln -s libcudnn.so.5 libcudnn.so 

   設置環境變量，終端輸入 

   sudo gedit /etc/profile 

   在末尾加入 

   PATH=/usr/local/cuda/bin:$PATH 
   export PATH 

   保存後，建立連接文件 

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

   按a進入插入模式，增長下面一行 

   /usr/local/cuda/lib64 

   按esc退出插入模式，按:wq保存退出 
   最後在終端輸入sudo ldconfig使連接生效 
   
   

5. cuda Samples測試

   打開CUDA 8.0 Samples默認安裝路徑，終端輸入 
   cd /home/yt/NVIDIA_CUDA-8.0_Samples （yt是個人用戶名） 
   sudo make all -j4 （4核） 
   出現「unsupported GNU version! gcc versions later than 5.3 are not supported!」的錯誤，這是因爲GCC版本太高，在終端輸入 
   cd /usr/local/cuda-8.0/include 
sudo cp host_config.h host_config.h.bak 
sudo gedit host_config.h 
   ctrl+f尋找有「5.3」的地方，只有一處，以下 
   # if __GNUC__ > 5 || (__GNUC__ == 5 && __GNUC_MINOR__ > 3) 
#error -- unsupported GNU version! gcc versions later than 5.3 are not supported! 
   將兩個5改爲6，即 
   #if __GNUC__ > 6 || (__GNUC__ == 6 && __GNUC_MINOR__ > 3) 
   保存退出，繼續在終端輸入 
   cd /home/yt/NVIDIA_CUDA-8.0_Samples （yt是個人用戶名） 
   sudo make all -j4 （4核） 
   完成後繼續向終端輸入 
   cd bin/x86_64/linux/release 
./deviceQuery 
   完成以後出現以下圖所示，表示成功安裝cuda 
   

---

3、依賴包安裝

1. sudo apt-get install build-essential #必要的編譯工具依賴

2. sudo apt-get install libprotobuf-dev libleveldb-dev libsnappy-dev libopencv-dev libhdf5-serial-dev protobuf-compiler

3. sudo apt-get install --no-install-recommends libboost-all-dev

4. sudo apt-get install libatlas-base-dev

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

---

四

1. 安裝python的pip和easy_install，方便安裝軟件包

   終端輸入 
   cd 
wget --no-check-certificate https://bootstrap.pypa.io/ez_setup.py 
sudo python ez_setup.py --insecure 
wget https://bootstrap.pypa.io/get-pip.py 
sudo python get-pip.py

---

五

1. 安裝科學計算和python所需的部分庫

   終端輸入 
   sudo apt-get install libblas-dev liblapack-dev libatlas-base-dev gfortran python-numpy

---

六

1. 安裝git，拉取源碼

   終端輸入 
   sudo apt-get install git 
git clone https://github.com/BVLC/caffe.git

---

七

1. 安裝python依賴

   終端輸入 
   sudo apt-get install python-pip 安裝pip 
   cd /home/yt/caffe/python
   sudo su 
for req in $(cat "requirements.txt"); do pip install -i https://pypi.tuna.tsinghua.edu.cn/simple $req; done
   按Ctrl+D退出sudo su模式

---

8、編譯caffe(暫不對matlab說明)

1. 終端輸入 
   cd /home/yt/caffe 
cp Makefile.config.example Makefile.config 
gedit Makefile.config

   ①將USE_CUDNN := 1取消註釋，

   ②INCLUDE_DIRS := $(PYTHON_INCLUDE) /usr/local/include後面打上一個空格 而後添加/usr/include/hdf5/serial若是沒有這一句可能會報一個找不到hdf5.h的錯誤

2. 終端輸入 
   make all -j4 
   make過程當中出現找不到lhdf5_hl和lhdf5的錯誤， 
   解決方案： 
   在計算機中搜索libhdf5_serial.so.10.1.0，找到後右鍵點擊打開項目位置 
   該目錄下空白處右鍵點擊在終端打開，打開新終端輸入 
   sudo ln libhdf5_serial.so.10.1.0 libhdf5.so 
sudo ln libhdf5_serial_hl.so.10.0.2 libhdf5_hl.so 
   最後在終端輸入sudo ldconfig使連接生效 
   原終端中輸入make clean清除第一次編譯結果 
   再次輸入make all -j4從新編譯

3. 終端輸入 

   make test -j4 
   make runtest -j4 
   make pycaffe -j4 
   make distribute   #生成發佈安裝包

    

4. 測試python，終端輸入 
   cd /home/yt/caffe/python 
python 
import caffe 
   若是不報錯就說明編譯成功

---

9、mnist測試

1. 下載mnist數據集，終端輸入 
   cd /home/yt/caffe/data/mnist/
./get_mnist.sh 獲取mnist數據集 
   在/home/yt/caffe/data/mnist/目錄下會多出訓練集圖片、訓練集標籤、測試集圖片和測試集標籤等4個文件

2. mnist數據格式轉換，終端輸入 
   cd /home/yt/caffe/
./examples/mnist/create_mnist.sh
   必需要在第一行以後運行第二行，即必需要在caffe根目錄下運行create_mnist.sh 
   此時在/caffe/examples/mnist/目錄下生成mnist_test_lmdb和mnist_train_lmdb兩個LMDB格式的訓練集和測試集

3. LeNet-5模型描述在/caffe/examples/mnist/lenet_train_test.prototxt

4. Solver配置文件在/caffe/examples/mnist/lenet_solver.prototxt

5. 訓練mnist，執行文件在/caffe/examples/mnist/train_lenet.sh
   終端輸入 
   cd /home/yt/caffe/
./examples/mnist/train_lenet.sh
   測試結果以下 
   

 

10、安裝theano

一、直接輸入命令：

sudo pip install theano 

二、配置參數文件：.theanorc

sudo gedit ~/.theanorc

[global]  
floatX=float32  
device=gpu  
base_compiledir=~/external/.theano/  
allow_gc=False  
warn_float64=warn  
[mode]=FAST_RUN  
  
[nvcc]  
fastmath=True  
  
[cuda]  
root=/usr/local/cuda 

三、運行測試例子：

from theano import function, config, shared, sandbox  
import theano.tensor as T  
import numpy  
import time  
  
vlen = 10 * 30 * 768  # 10 x #cores x # threads per core  
iters = 1000  
  
rng = numpy.random.RandomState(22)  
x = shared(numpy.asarray(rng.rand(vlen), config.floatX))  
f = function([], T.exp(x))  
print(f.maker.fgraph.toposort())  
t0 = time.time()  
for i in range(iters):  
    r = f()  
t1 = time.time()  
print("Looping %d times took %f seconds" % (iters, t1 - t0))  
print("Result is %s" % (r,))  
if numpy.any([isinstance(x.op, T.Elemwise) for x in f.maker.fgraph.toposort()]):  
    print('Used the cpu')  
else:  
    print('Used the gpu') 

10、安裝tensosrflow

一、安裝編譯工具Bazel

echo "deb [arch=amd64] http://storage.googleapis.com/bazel-apt stable jdk1.8" | sudo tee /etc/apt/sources.list.d/bazel.list 

curl https://storage.googleapis.com/bazel-apt/doc/apt-key.pub.gpg | sudo apt-key add -  

sudo apt-get update && sudo apt-get install bazel  

sudo apt-get upgrade bazel

二、下載tensorflow並編譯

git clone https://github.com/tensorflow/tensorflow
cd tensorflow
git checkout Branch # where Branch is the desired branch
git checkout r1.0

 

sudo apt-get install python-numpy python-dev python-pip python-wheel
sudo apt-get install python3-numpy python3-dev python3-pip python3-wheel

 

sudo apt-get install libcupti-dev 

 

./configure  

$ ./configure # 如下是一個例子
Please specify the location of python. [Default is /usr/bin/python]: y
Invalid python path. y cannot be found
Please specify the location of python. [Default is /usr/bin/python]: 
Please specify optimization flags to use during compilation when bazel option "--config=opt" is specified [Default is -march=native]: 
Do you wish to use jemalloc as the malloc implementation? [Y/n] y
jemalloc enabled
Do you wish to build TensorFlow with Google Cloud Platform support? [y/N] n
No Google Cloud Platform support will be enabled for TensorFlow
Do you wish to build TensorFlow with Hadoop File System support? [y/N] y
Hadoop File System support will be enabled for TensorFlow
Do you wish to build TensorFlow with the XLA just-in-time compiler (experimental)? [y/N] n
No XLA JIT support will be enabled for TensorFlow
Found possible Python library paths:
  /usr/local/lib/python2.7/dist-packages
  /usr/lib/python2.7/dist-packages
Please input the desired Python library path to use.  Default is [/usr/local/lib/python2.7/dist-packages]

Using python library path: /usr/local/lib/python2.7/dist-packages
Do you wish to build TensorFlow with OpenCL support? [y/N] n
No OpenCL support will be enabled for TensorFlow
Do you wish to build TensorFlow with CUDA support? [y/N] y
CUDA support will be enabled for TensorFlow
Please specify which gcc should be used by nvcc as the host compiler. [Default is /usr/bin/gcc]: 
Please specify the CUDA SDK version you want to use, e.g. 7.0. [Leave empty to use system default]: 8.0
Please specify the location where CUDA 8.0 toolkit is installed. Refer to README.md for more details. [Default is /usr/local/cuda]: 
Please specify the Cudnn version you want to use. [Leave empty to use system default]: 5
Please specify the location where cuDNN 5 library is installed. Refer to README.md for more details. [Default is /usr/local/cuda]: 
Please specify a list of comma-separated Cuda compute capabilities you want to build with.
You can find the compute capability of your device at: https://developer.nvidia.com/cuda-gpus.
Please note that each additional compute capability significantly increases your build time and binary size.
[Default is: "3.5,5.2"]: 6.1
INFO: Starting clean (this may take a while). Consider using --expunge_async if the clean takes more than several minutes.
........
INFO: All external dependencies fetched successfully.
Configuration finished

 編譯

bazel build --config=opt --config=cuda //tensorflow/tools/pip_package:build_pip_package --cxxopt="-D_GLIBCXX_USE_CXX11_ABI=0"

 

bazel-bin/tensorflow/tools/pip_package/build_pip_package /tmp/tensorflow_pkg 

檢查tmp文件夾下生成的whl文件名

sudo pip install /tmp/tensorflow_pkg/ tensorflow-1.0.1-cp27-cp27mu-linux_x86_64.whl

三、測試

python
import tensorflow as tf
sess = tf.Session()