安裝nvidia驅動和CUDA

1. 下載驅動和CUDA安裝包，在官網下載對應版本就行
2. sudo apt-get install linux-headers-$(uname -r) 或者 linux-headers-generic.否則直接安裝會報錯 kernel not found
3. 安裝 nvidia 驅動，一路accept和yes
4. 安裝 CUDA，一路yes。安裝路徑：/usr/local/cuda-8.0/。是否安裝推薦的驅動, no
   最後顯示類似下面Summary內容，表示安裝成功。
5. 在/etc/profile中新增：

export PATH=/usr/local/cuda-8.0/lib64:$PATH
 

export LD_LIBRARY_PATH=/usr/local/cuda/lib64:$LD_LIBRARY_PATH

儲存後

# source /etc/profile
# nvcc -V  檢查CUDA
# apt-get install cmake 安裝cmake
# cd  /usr/local/cuda-8.0/samples
# make  測試CUDA

測試時間較長，一段沒有error即可中止

===========
= Summary =
===========

Driver:   Not Selected
Toolkit:  Installed in /usr/local/cuda-8.0
 

Samples:  Installed in /storage/installers/cuda_samples, but missing recommended libraries

Please make sure that
 -   PATH includes /usr/local/cuda-8.0/bin
 -   LD_LIBRARY_PATH includes /usr/local/cuda-8.0/lib64, or, add /usr/local/cuda-8.0/lib64 to /etc/ld.so.conf and run ldconfig as root

To uninstall the CUDA Toolkit, run
 
 the uninstall script in /usr/local/cuda-8.0/bin

Please see CUDA_Installation_Guide_Linux.pdf in /usr/local/cuda-8.0/doc/pdf for detailed information on setting up CUDA.

***WARNING: Incomplete installation! This installation did not install the CUDA Driver. A driver of version at least 361.00 is required for CUDA 8.0 functionality to work.
To install the driver using this installer, run the following command, replacing <CudaInstaller> with the name of this run file:
    sudo <CudaInstaller>.run -silent -driver

Logfile is /tmp/cuda_install_16018.log

安裝CUDNN

1. 下載cudnn-8.0-linux-x64-v5.1.tgz
2. tar -xvzf cudnn-8.0-linux-x64-v5.1.tgz 解壓完會有一個cuda資料夾

# cd cuda
# cp  include/cudnn.h
 /usr/local/cuda/include
# cp  lib64/libcudnn.*   /usr/local/cuda/lib64

cuDNN安裝完成！
有的部落格說要建立軟連結，但是我沒有做，步驟如下：

# cd  /usr/local/cuda/lib64 
# rm  -rf  libcudnn.so  libcudnn.so.5
# ln  -s  libcudnn.5.1.3  libcudnn.so.5
# ln  -s  libcudnn.so.5  libcudnn.so

首先安裝anaconda, 因為很多python庫都包含在裡面了,一次性安裝很方便

安裝anaconda

1. 從官網下載最新的anaconda安裝包，我下的是Anaconda2-4.2.0
2. bash Anaconda2-4.2.0-Linux-x86_64.sh
3. PREFIX=/usr/share/anaconda2
4. # vim /etc/profile (新增環境變數)
   export PATH=$PATH:/usr/share/anaconda2/bin
   再source /etc/profile生效
   修改映象檔案，使得系統預設python為anaconda中的python

# mv /usr/bin/python  /usr/bin/python_bk
# ln -s /usr/share/anaconda2/bin/python  /usr/bin/python

安裝h5py

conda install h5py #注意必須先安裝anaconda2
這時會提示升級anaconda,yes即可

安裝tensorflow

ln -s /usr/anaconda2/bin/pip /usr/bin/pip 建立軟連線
github下載 tensorflow_gpu-0.12.0rc0-cp27-none-linux-x86_64.whl
pip install tensorflow_gpu-0.12.0rc0-cp27-none-linux-x86_64.whl
安裝完成。測試：
python && import tensorflow 測試tensorflow
或者：

# cd  /usr/share/anaconda2/lib/python2.7/site-packages/tensorflow/models/image/mnist
# CUDA_VISIBLE_DEVICES = 0(選擇顯示卡)  python convolutional.py

開頭出現以下字樣表示安裝成功：

Python 2.7.12 |Anaconda 4.2.0 (64-bit)| (default, Jul  2 2016, 17:42:40) 
[GCC 4.4.7 20120313 (Red Hat 4.4.7-1)] on linux2
Type "help", "copyright", "credits" or "license" for more information.
Anaconda is brought to you by Continuum Analytics.
Please check out: http://continuum.io/thanks and https://anaconda.org
>>> import tensorflow as tf
I tensorflow/stream_executor/dso_loader.cc:128] successfully opened CUDA library libcublas.so locally
I tensorflow/stream_executor/dso_loader.cc:128] successfully opened CUDA library libcudnn.so locally
I tensorflow/stream_executor/dso_loader.cc:128] successfully opened CUDA library libcufft.so locally
I tensorflow/stream_executor/dso_loader.cc:128] successfully opened CUDA library libcuda.so.1 locally
I tensorflow/stream_executor/dso_loader.cc:128] successfully opened CUDA library libcurand.so locally