重新在臺式機上學習深度學習，Ubuntu18.04 + Tensorflow-gpu + cuda8.0 + cuDNN6.0

一、準備工作（有一些不是必需的）：

1.安裝pip：

開啟終端輸入命令：sudo apt-get install python-pip python-dev

2.安裝vim：

sudo apt-get install vim-gtk

在命令列下，輸入命令：sudo vim /etc/vim/vimrc    必須加上sudo，否則你是沒有許可權編輯vimrc的。

在這個檔案中，會有這麼一句： syntax on   意思是語法高亮，如果您的被註釋掉了，請“讓它出來”。

請在您的VIM的最後一行，輸入他們，可以讓您的VIM變得更漂亮、舒服。

set nu                           // 在左側行號

set tabstop                  //tab 長度設定為 4

set nobackup               //覆蓋檔案時不備份

set cursorline               //突出顯示當前行

set ruler                       //在右下角顯示游標位置的狀態行

set autoindent             //自動縮排

二、ubuntu安裝GPU顯示卡驅動

開啟終端：sudo apt-get update

選擇系統設定→軟體更新→附加驅動→選擇nvidia最新驅動→應用更改.（如果沒有，去官網找對應的驅動安裝即可）

驗證安裝成功：軟體連找到 NVIDIA X Sever Settings

驅動安裝成功！

三、安裝Tensorflow依賴的編譯工具bazel

1、安裝bazel前，需先安裝JDK8

sudo apt-get install software-properties-common

sudo add-apt-repository ppa:webupd8team/java

sudo apt-get update

sudo apt-get install oracle-java8-installer

安裝完成後 驗證java版本：java -version

2、安裝bazel

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

sudo apt install curl

安裝和更新Bazel

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

sudo apt-get upgrade bazel 

三、安裝cuda 8.0（先看本章 4、驗證cuda8是否安裝成功）

1、cuda 8.0 下載地址：https://developer.nvidia.com/cuda-80-ga2-download-archive

驗證主機是否滿足cuda8.0安裝的要求

lspci | grep -i nvidia

uname -m && /etc/*release
gcc --version
uname -r
sudoapt-get install linux-headers-$(uname -r)

sudo sh cuda_8.0.27_linux.run

2、新增環境變數

cd ~

vim   .bashrc

末尾新增

export PATH=/usr/local/cuda-8.0/bin:$PATH

export LD_LIBRARY_PATH=/usr/local/cuda-8.0/lib64:/usr/local/cuda-8.0/extras/    CUPTI/lib64:$LD_LIBRARY_PATH

然後 source ~/.bashrc

3、nvcc -V

如果沒有nvcc則安裝 sudo apt install nvidia-cuda-toolkit

nvcc -V

nvidia-smi

4、驗證cuda8是否安裝成功：

進入NVIDIA_CUDA-8.0_Samples目錄，執行：$ make

出現錯誤！！！

原因：由於cuda8.0不支援gcc 5.0以上的編譯器，因此需要降級，把編譯器版本降到4.8

5、g++/gcc降級到4.8（建議降級到4.9，我這裡到4.9失敗所以用4.8）

sudo apt-get -y update

sudo apt-get install -y gcc-4.9

sudo apt-get install -y g++-4.9

cd /usr/bin

sudo rm gcc

sudo ln -s gcc-4.8 gcc

sudo rm g++

sudo ln -s g++-4.8 g++

6、CUDA8.0安裝成功測試

進入NVIDIA_CUDA-8.0_Samples目錄，執行：$ sudo make

進入1_Utilities目錄，執行：$ ./deviceQuery/deviceQuery ，結果如下：

執行：$ nvcc -V，結果如下：

驗證CUDA8.0安裝成功，demo也跑通了。

四、安裝cuDNN6.0（因為5.1在編譯步驟不通過）

下載地址6.0： https://developer.nvidia.com/rdp/cudnn-archive

開啟檔案所在資料夾，當前位置開啟終端

tar -xvzf cudnn-8.0-linux-x64-v6.0.tgz
sudo cp -P cuda/include/cudnn.h /usr/local/cuda-8.0/include
sudo cp -P cuda/lib64/libcudnn* /usr/local/cuda-8.0/lib64

sudo chmod a+r /usr/local/cuda-8.0/include/cudnn.h /usr/local/cuda-8.0/lib64/libcudnn*

配置環境變數；

sudo gedit ~/.bashrc

新增：

export LD_LIBRARY_PATH=”$LD_LIBRARY_PATH:/usr/local/cuda/lib64:/usr/local/cuda/extras/CUPTI/lib64”
export CUDA_HOME=/usr/local/cuda

export PATH="$CUDA_HOME/bin:$PATH"

再執行：source ~/.bashrc

配置完成！

五、Tensorflow（GPU）安裝

方法一：pip 二進位制檔案安裝

pip install https://storage.googleapis.com/tensorflow/linux/gpu/tensorflow_gpu-1.2.0-cp27-none-linux_x86_64.whl

方法二：原檔案安裝

1、安裝TF依賴工具包

sudo apt-get install python-numpy swigpython-dev python-wheel

2、下載最新原始碼

sudo apt-get install git

3、執行configure指令碼配置環境資訊

cd tensorflow

./configure

並不是預設選擇Y就可以了，巨大部分是選擇N，具體他有些選項不會出現，需要另外查詢是不是需要選擇y/n，計提問題具體分析，下面是我的選項。

[email protected]:~/tensorflow$ ./configure
WARNING: Running Bazel server needs to be killed, because the startup options are different.
You have bazel 0.14.1 installed.
Please specify the location of python. [Default is /usr/bin/python]: 


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]
/usr/bin/python
Do you wish to build TensorFlow with jemalloc as malloc support? [Y/n]: n
No jemalloc as malloc support will be enabled for TensorFlow.

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]: n
No Hadoop File System support will be enabled for TensorFlow.

Do you wish to build TensorFlow with Amazon S3 File System support? [Y/n]: n
No Amazon S3 File System support will be enabled for TensorFlow.

Do you wish to build TensorFlow with Apache Kafka Platform support? [Y/n]: n
No Apache Kafka Platform support will be enabled for TensorFlow.

Do you wish to build TensorFlow with XLA JIT support? [y/N]: n
No XLA JIT support will be enabled for TensorFlow.

Do you wish to build TensorFlow with GDR support? [y/N]: n
No GDR support will be enabled for TensorFlow.

Do you wish to build TensorFlow with VERBS support? [y/N]: n
No VERBS support will be enabled for TensorFlow.

Do you wish to build TensorFlow with OpenCL SYCL support? [y/N]: n
No OpenCL SYCL 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 the CUDA SDK version you want to use. [Leave empty to default to CUDA 9.0]: 8


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 default to cuDNN 7.0]: 6


Please specify the location where cuDNN 6 library is installed. Refer to README.md for more details. [Default is /usr/local/cuda]:


Do you wish to build TensorFlow with TensorRT support? [y/N]: n
No TensorRT support will be enabled for TensorFlow.

Please specify the NCCL version you want to use. [Leave empty to default to NCCL 1.3]: 


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: 6.1]


Do you want to use clang as CUDA compiler? [y/N]: n
nvcc will be used as CUDA compiler.

Please specify which gcc should be used by nvcc as the host compiler. [Default is /usr/bin/gcc]: 


Do you wish to build TensorFlow with MPI support? [y/N]: n
No MPI support will be enabled for TensorFlow.

Please specify optimization flags to use during compilation when bazel option "--config=opt" is specified [Default is -march=native]: 


Would you like to interactively configure ./WORKSPACE for Android builds? [y/N]: n
Not configuring the WORKSPACE for Android builds.

Preconfigured Bazel build configs. You can use any of the below by adding "--config=<>" to your build command. See tools/bazel.rc for more details.
	--config=mkl         	# Build with MKL support.
	--config=monolithic  	# Config for mostly static monolithic build.
Configuration finished

4、通過bazel來編譯pip的安裝包，然後通過pip安裝

bazel build -c opt --copt=-mavx --copt=-mavx2 --copt=-mfma --copt=-mfpmath=both --copt=-msse4.2 --config=cuda -k //tensorflow/tools/pip_package:build_pip_package

如果出現錯誤：

Cuda Configuration Error: Cannot find libdevice.10.bc under /usr/local/cuda-8.0
WARNING: Target pattern parsing failed.
ERROR: no such package '@local_config_cuda//crosstool': Traceback (most recent call last):

解決方案：將/usr/local/cuda-8.0/nvvm/libdevice/libdevice.compute_50.10.bc改為libdevice.10.bc，並複製一份至/usr/local/cuda-8.0/

再執行上面的命令，上面的過程比較耗時，結束後繼續執行。

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

sudo pip install/tmp/tensorflow_pkg/tensorflow-1.2.0rc2-cp27-cp27mu-linux_x86_64.whl

第一個命令中 --config=cuda引數為對GPU的支援，如何不需要支援GPU，就不需要這個引數。

 5\測試：

import tensorflow as tf

a=tf.constant([1.0,2.0,3.0],shape=[3],name='a')

b=tf.constant([1.0,2.0,3.0],shape=[3],name='b')

c=a+b

sess=tf.Session(config=tf.ConfigProto(log_device_placement=True))

print sess.run(c)