Install nVidia graphics card driver and cuda/cudnn in ubuntu 16.04.
Recommended new version installation tutorial
Http://blog.csdn.net/chenhaifeng2016/article/details/78874883
To install the deep learning framework, you must use cuda/cudnn (GPU) to accelerate computing. To install cuda/cudnn, you must first install the nvidia graphics card driver.
During the installation process, I encountered a driver conflict and had to re-install the operating system.
The information on the internet is messy, and many of them are reprinted. Some versions of the operating system are too low. Now I write down the entire process for reference by people who encounter the same problem.
By default, the third-party open-source driver nouveau is installed in ubuntu 16.04. to install the nvidia graphics driver, you must first disable nouveau. Otherwise, the nvidia graphics driver cannot be installed due to a conflict.
Edit the file blacklist. conf
Sudo vim/etc/modprobe. d/blacklist. conf
Insert the following two lines at the end of the file:
Blacklist nouveau
Options nouveau modeset = 0
Update System
Sudo update-initramfs-u
Restart the system (Be sure to restart)
Verify that nouveau is disabled
Lsmod | grep nouveau
No information is displayed, indicating that nouveau has been disabled. Next, you can install the nvidia graphics driver.
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. Otherwise, loop logon may occur even if the installation is successful (at least I have encountered such a situation ).
Do not install the/usr/local/cuda symbolic connection (this step is optional)
Sudo service lightdm start
Finally copy the header file of cudnn to/usr/local/cuda-8.0/include, library file copy to/usr/local/cuda-8.0/lib64
Set the environment variable file. bashrc or profile
Export PATH =/usr/local/The 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 is complete.
Next, you can install the tensorflow, caffe, and mxnet deep learning frameworks.
-
Top
-
1
-
Step on
-
0
View comments