Machine Learning Environment configuration Series 1 Cuda

The environment configured in this article is redhat6.9 + cuda10.0 + cudnn7.3.1 + anaonda6.7 + theano1.0.0 + keras2.2.0 + jupyter remote, with Cuda version 10.0.

Step 1: before installing Cuda:

1. Verify if GPU is installed

 
$ Lspci | grep-I NVIDIA
 

2. Check the RedHat version.

 
$ Uname-M & CAT/etc/* release
 

3. After the test is completed, download Cuda from the NVIDIA website based on the version. The installed operating system is redhat6.9 and the video card is Quadro m6000.

1. Download the Cuda version of redhat6 from the NVIDIA website (this download is the run file)
2. Log on to red hat with the created user (such as the mlstudy user;
3. Stop running nouveau before installation

• • Create File for/etc/modprobe. d/blacklist-nouveau.conf content:

Blacklist nouveauoptions nouveau modeset = 0
 

• • Restart the system (reboot) and enter reboot into text mode (runlevel 3). You can also directly modify the/etc/inittab file and restart the system.

 
VI/etc/inittab
File Content
 

0-halt (do not set initdefault to this)
#1-Single User Mode
#2-multiuser, without NFS (the same as 3, if you do not have networking)
#3-full multiuser Mode
#4-unused
#5-X11
#6-Reboot (do not set initdefault to this)
#
ID: 5: initdefault:

Modify ID: 3: initdefault:

• • Run the Cuda installation command

 
$ Sudo sh Cuda _ <version> _ Linux. Run
 

Follow the prompts to install Cuda.

Machine Learning Environment configuration Series 1 Cuda