caffe安装（1）ubuntu16.04+显卡驱动+cuda8.0
0. 安装ubuntu16.04系统
F2进入BIOS, 设置U盘启动（原来已装了ubuntu的要取消快速启动)
按照提示安装ubuntu16.04想，选择英文版本，中午版本的路径一看就晕
安装完会发现分辨率不对，而且改不了，然而安装完显卡驱动就好啦，所以先不管
1.安装显卡驱动
上nividia官网下载相应版本
下载 后右击，修改属性为可执行文件
sudo service lightdm stop
sudo ./NVIDIA-Linux-x86_64-375.20.run
按照提示安装
sudo service lightdm restart安装完看显卡是否安装成功，可以看到nvidia-settings为安装成功
nvidia-settings查看nividia显卡信息: nvidia-smi
Sat Feb 4 14:11:46 2017
+-----------------------------------------------------------------------------+
| NVIDIA-SMI 367.48 Driver Version: 367.48 |
|-------------------------------+----------------------+----------------------+
| GPU Name Persistence-M| Bus-Id Disp.A | Volatile Uncorr. ECC |
| Fan Temp Perf Pwr:Usage/Cap| Memory-Usage | GPU-Util Compute M. |
|===============================+======================+======================|
| 0 TITAN X (Pascal) Off | 0000:02:00.0 On | N/A |
| 23% 24C P8 13W / 250W | 191MiB / 12186MiB | 2% Default |
+-------------------------------+----------------------+----------------------+
+-----------------------------------------------------------------------------+
| Processes: GPU Memory |
| GPU PID Type Process name Usage |
|=============================================================================|
| 0 4578 G /usr/lib/xorg/Xorg 148MiB |
| 0 5329 G compiz 41MiB |
+-----------------------------------------------------------------------------+卸载的话
查看和nvidia相关的dpkg -l | grep -i nvidia
sudo apt-get remove --purge nvidia-*
sudo apt-get install ubuntu-desktop2. 安装CUDA
2.1 下载：(https://developer.nvidia.com/cuda-downloads)
选择对应的版本
2.2 修改为可执行文件
sudo chmod +x cuda_8.0.44_linux.run2.3 安装依赖库，这些是安装cuda-sample需要的
sudo apt-get install freeglut3-dev
sudo apt-get install build-essential
sudo apt-get installlibx11-dev
sudo apt-get install libxmu-dev
sudo apt-get install libxi-dev
sudo apt-get installlibglu1-mesa-dev2.4 执行
sudo ./cuda_8.0.44_linux.run
一开始是说明，ctrl+c跳过那些文字，accept接受要求
Do you accept the previously read EULA?
accept/decline/quit: accept
Install NVIDIA Accelerated Graphics Driver for Linux-x86_64 367.48?
(y)es/(n)o/(q)uit: n
Install the CUDA 8.0 Toolkit?
(y)es/(n)o/(q)uit: y
Enter Toolkit Location
[ default is /usr/local/cuda-8.0 ]:
Do you want to install a symbolic link at /usr/local/cuda?
(y)es/(n)o/(q)uit: y
Install the CUDA 8.0 Samples?
(y)es/(n)o/(q)uit: y
Enter CUDA Samples Location
[ default is /home/kinny ]:
一定要注意提示安不安装 NVIDIA Accelerated Graphics Driver for Linux-x86_64 367.48? 要选择不，我因为这个版本可能跟你自己的不一样，安装了重启后会进不去桌面
2.5 添加PATH路径,lib路径
sudo gedit /etc/profile
添加
export PATH=/usr/local/cuda-8.0/bin:$PATH
export LD_LIBRARY_PATH=/usr/local/cuda-8.0/lib64:$LD_LIBRARY_PATH保存，执行source /etc/profile
2.6 验证安装成功
sudo reboot
cd ~/NVIDIA_CUDA-8.0_Samples
make
cd ~/NVIDIA_CUDA-8.0_Samples/bin/x86_64/linux/release
./deviceQuery显示 Result = PASS
3. 安装CuDNN
官网下载CuDNN，需要邮箱注册，填写信息
sudo cp cuda/include/cudnn.h /usr/local/cuda/include
sudo cp cuda/lib64/libcudnn* /usr/local/cuda/lib64
sudo chmod a+r /usr/local/cuda/include/cudnn.h /usr/local/cuda/lib64/libcudnn*4.安装NCCL （支持多显卡）
cd nccl
make CUDA_HOME=/usr/local/cuda test
make install
echo 'export LD_LIBRARY_PATH=/usr/local/lib:$LD_LIBRARY_PATH'>>~/.bashrc
3) 测试
cd nccl/build/test/single
./all_reduce_test 1000000
参考：Ubuntu 16.04安装配置TensorFlow GPU版本