wangruohui

In this article, I will share some of my experience on installing NVIDIA driver and CUDA on Linux OS. Here I mainly use Ubuntu as example. Comments for CentOS/Fedora are also provided as much as I can.

Table of Contents

• Install NVIDIA Graphics Driver via apt-get
• Install NVIDIA Graphics Driver via runfile
  • Remove Previous Installations (Important)
  • Download the Driver
  • Install Dependencies
  • Creat Blacklist for Nouveau Driver

• Stop lightdm/gdm/kdm

idleberg

Open in Atom

• Open Automator
• Create a new Service
• Set “Service receives selected” to files or folders in any application
• Add a Run Shell Script action
• Set the script action to /usr/local/bin/atom -n "$@"
• Set “Pass input” to as arguments
• Save as Open in Atom

beeman

# Stop all containers
docker stop `docker ps -qa`

# Remove all containers
docker rm `docker ps -qa`

# Remove all images
docker rmi -f `docker images -qa `

# Remove all volumes

alexlee-gk

This was tested on a ThinkPad P70 laptop with an Intel integrated graphics and an NVIDIA GPU:

lspci | egrep 'VGA|3D'
00:02.0 VGA compatible controller: Intel Corporation Device 191b (rev 06)
01:00.0 VGA compatible controller: NVIDIA Corporation GM204GLM [Quadro M3000M] (rev a1)

A reason to use the integrated graphics for display is if installing the NVIDIA drivers causes the display to stop working properly. In my case, Ubuntu would get stuck in a login loop after installing the NVIDIA drivers. This happened regardless if I installed the drivers from the "Additional Drivers" tab in "System Settings" or the ppa:graphics-drivers/ppa in the command-line.

sergeyprokudin

import numpy as np


def count_conv_params_flops(conv_layer, verbose=1):
    # out shape is  n_cells_dim1 * (n_cells_dim2 * n_cells_dim3)
    out_shape = conv_layer.output.shape.as_list()
    n_cells_total = np.prod(out_shape[1:-1])

    n_conv_params_total = conv_layer.count_params()