1.gcc/g++降级

Ubuntu18.04自带的gcc/g++是7.0版本的，但cuda不支持这么高版本，我们需要安装4.8版本。

1.下载安装4.8版本的gcc/g++

sudo apt-get install gcc-4.8 
sudo apt-get install g++-4.8

2.让gcc软连接至4.8版本的gcc，g++软连接至4.5版本的g++

装完后进入到/usr/bin目录下

sudo mv gcc gcc.bak #备份 
sudo ln -s gcc-4.8 gcc #重新链接

对g++也进行同样的操作

3.再查看gcc和g++版本号

gcc -v g++ -v

均显示gcc version 4.8 ，说明gcc 48.8安装成功。

2.安装GPU（针对ubuntu18.04）

安装所有推荐的驱动：

sudo ubuntu-drivers autoinstall

3.cuda安装

1.下载cuda安装包，我们安装9.0版本，地址：CUDA

将上面的选项框填好后，会弹出下面的选项框

下载CUDA9.0和全部四个补丁（这几个补丁不安装的话，cuda在使用过程中可能出现各种bug），他们会被保存到该用户目录下的 Downloads 文件夹下。

2.安装cuda

进入Downloads/后，输入：

sh cuda_9.0.176_384.81_linux.run 
sh cuda_9.0.176.1_linux.run 
sh cuda_9.0.176.2_linux.run 
sh cuda_9.0.176.3_linux.run 
sh cuda_9.0.176.4_linux.run

安装这五个文件

注意：执行，如果有安装了显卡驱动的，注意在提问是否安装显卡驱动时选择no（因为在前边已经装过了）,其他选择默认路径或者yes即可。

注：我们也可以不在usr/local下安装cuda,而在用户根目录下安装：

然后就可以了

3.添加访问cuda的路径

安装完毕之后，将以下两条加入.barshrc文件中.

sudo vim ~/.barshrc
export PATH=/usr/local/cuda-9.0/bin${PATH:+:$PATH}} 
export LD_LIBRARY_PATH=/usr/local/cuda-9.0/lib64${LD_LIBRARY_PATH:+:${LD_LIBRARY_PATH}}

下边的这两条和上边的等价（暂时没搞清楚为什么）：

sudo vim ~/.barshrc
export PATH=/usr/local/cuda-9.0/bin:$PATH 
export LD_LIBRARY_PATH=/usr/local/cuda-9.0/lib64:$LD_LIBRARY_PATH

4.CUDNN安装

CUDNN需要注册，我们安装7.1.3版本

注册完以后，出现下面的界面：

下载结束后，将压缩包进行解压缩。

然后，在Downloads/文件夹内输入如下命令，将CUDNN拷贝至CUDA的目录中（注，CUDNN无需安装）：

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*

5.cuda测试（看cuda是否安装好）

进入samples文件夹，一般在home目录下

cd ~/NVIDIA_CUDA-9.1_Samples/
make

编译完成后，进入：

cd ./bin/x86_64/linux/release

使用deviceQuery 或 bandwidthTest测试

$ ./deviceQuery

./deviceQuery Starting...

 CUDA Device Query (Runtime API) version (CUDART static linking)

Detected 1 CUDA Capable device(s)

Device 0: "GeForce GTX 960M"
  CUDA Driver Version / Runtime Version          9.1 / 9.1
  CUDA Capability Major/Minor version number:    5.0
  Total amount of global memory:                 2004 MBytes (2101870592 bytes)
  ( 5) Multiprocessors, (128) CUDA Cores/MP:     640 CUDA Cores
  GPU Max Clock rate:                            1176 MHz (1.18 GHz)
  Memory Clock rate:                             2505 Mhz
  Memory Bus Width:                              128-bit
  L2 Cache Size:                                 2097152 bytes
  Maximum Texture Dimension Size (x,y,z)         1D=(65536), 2D=(65536, 65536), 3D=(4096, 4096, 4096)
  Maximum Layered 1D Texture Size, (num) layers  1D=(16384), 2048 layers
  Maximum Layered 2D Texture Size, (num) layers  2D=(16384, 16384), 2048 layers
  Total amount of constant memory:               65536 bytes
  Total amount of shared memory per block:       49152 bytes
  Total number of registers available per block: 65536
  Warp size:                                     32
  Maximum number of threads per multiprocessor:  2048
  Maximum number of threads per block:           1024
  Max dimension size of a thread block (x,y,z): (1024, 1024, 64)
  Max dimension size of a grid size    (x,y,z): (2147483647, 65535, 65535)
  Maximum memory pitch:                          2147483647 bytes
  Texture alignment:                             512 bytes
  Concurrent copy and kernel execution:          Yes with 1 copy engine(s)
  Run time limit on kernels:                     Yes
  Integrated GPU sharing Host Memory:            No
  Support host page-locked memory mapping:       Yes
  Alignment requirement for Surfaces:            Yes
  Device has ECC support:                        Disabled
  Device supports Unified Addressing (UVA):      Yes
  Supports Cooperative Kernel Launch:            No
  Supports MultiDevice Co-op Kernel Launch:      No
  Device PCI Domain ID / Bus ID / location ID:   0 / 1 / 0
  Compute Mode:
     < Default (multiple host threads can use ::cudaSetDevice() with device simultaneously) >

deviceQuery, CUDA Driver = CUDART, CUDA Driver Version = 9.1, CUDA Runtime Version = 9.1, NumDevs = 1
Result = PASS

$ ./bandwidthTest

[CUDA Bandwidth Test] - Starting...
Running on...

 Device 0: GeForce GTX 960M
 Quick Mode

 Host to Device Bandwidth, 1 Device(s)
 PINNED Memory Transfers
   Transfer Size (Bytes)    Bandwidth(MB/s)
   33554432         12339.9

 Device to Host Bandwidth, 1 Device(s)
 PINNED Memory Transfers
   Transfer Size (Bytes)    Bandwidth(MB/s)
   33554432         11720.0

 Device to Device Bandwidth, 1 Device(s)
 PINNED Memory Transfers
   Transfer Size (Bytes)    Bandwidth(MB/s)
   33554432         65699.6

Result = PASS

NOTE: The CUDA Samples are not meant for performance measurements. Results may vary when GPU Boost is enabled.

在usr/local/cuda-9.0中输入nvcc -V

会出现：


yhao@yhao-X550VB:~$ nvcc -V

nvcc: NVIDIA (R) Cuda compiler driver

Copyright (c) 2005-2016 NVIDIA Corporation

Built on Tue_Jan_10_13:22:03_CST_2017

Cuda compilation tools, release 9.0, V8.0.61

6.tensorflow-gpu安装

在anaconda官网上下载anaconda并安装好后，输入：

pip install tensorflow-gpu==1.5.0

可以看到pip更新后，anaconda里边的库也会更新（如果是conda install 会出现问题，原因暂时没有搞清楚）

安装好后，可以在spyder-python中使用

import tensorflow as tf

有可能tensorflow在anaconda上不好使，这时候可以安装一个pycharm来运行tensorflow

7.cuda/cudnn版本与tensorflow版本对应关系

cuda/cudnn需要安装对应版本的tensorflow，不然会出现各种bug:

cuda8/cudnn5 --> tensorflow1.2及以下

cuda8/cudnn6 --> tensorflow1.3以及1.4

cuda9/cudnn7 --> tensorflow1.5及以上

Ubuntu 18.04 + Cuda + Cudnn + tensorflow-gpu