Ubuntu 16.04 installs Nvidia graphics driver and CUDA/CUDNN pit process

Recommended New Installation Tutorials

http://blog.csdn.net/chenhaifeng2016/article/details/78874883

The install Depth Learning framework requires the use of CUDA/CUDNN (GPU) to speed up computing, while installing CUDA/CUDNN requires Nvidia's graphics driver to be installed first.

I ran into a driver conflict during the installation, looping through the two issues so that I finally had to reinstall the operating system.

The information on the Internet is very messy, many are reproduced, some for the operating system version is too low, now I write down the whole process, for people who have the same problem for reference.

Ubuntu 16.04 defaults to install Third-party Open source driver nouveau, install NVIDIA graphics driver first need to disable Nouveau, otherwise encounter the problem of conflict, can not install Nvidia graphics driver.

Edit File blacklist.conf

sudo vim/etc/modprobe.d/blacklist.conf

Insert the following two lines in the last section of the file

Blacklist nouveau

Options Nouveau modeset=0

Update system

sudo update-initramfs-u

Reboot the system (be sure to reboot)

Verify Nouveau is disabled

Lsmod | grep nouveau

There is no information to show that Nouveau has been disabled, and then the NVIDIA graphics driver can be installed.

Press CTRL+ALT+F1 to enter text mode

sudo service LIGHTDM stop

sudo./cuda_8.0.61_375.26_linux.run

Do not install OpenGL, or even if the installation is successful, there will be a circular login (at least I encountered this situation).

Do not install/usr/local/cuda symbolic connections (optional)

sudo service LIGHTDM start

Finally, copy the CUDNN header files to the/usr/local/cuda-8.0/include below, and copy the library files to the/usr/local/cuda-8.0/lib64 below.

Sets the environment variable file. BASHRC or Profile

Export Path=/usr/local/cuda-8.0/bin: $PATH

Export ld_library_path=/usr/local/cuda-8.0/lib64: #LD_LIBRARY_PATH

Update environment variables

Souce/etc/profile

Installation complete.

Next you can install TensorFlow, Caffe, mxnet depth learning framework.