Mark Cotton mcotton

How to set up Ubuntu Server 20.04 with GPU to use TensorFlow and Pytorch via JupyterLab on Docker

Environment

Ubuntu Server 20.04
CUDA Driver 10.2
Docker 19.03
NVIDIA Container Toolkit 2.3.0
NVIDIA RTX 2080ti (or other supported GPUs)

Preparation

In open Ubuntu 18.04 machine click Parallels Actions -> "Install Parallels Tools"
A "Parallels Tools" CD will popup on your Ubuntu desktop.
Open it by double mouse click, copy all the content to a new, empty directory on a desktop, name it for e.g. "parallels_fixed"
Open terminal, change directory to parallels_fixed (cd ~/Desktop/parallels_fixed)
Make command line installer executable (chmod +x install)
Change directory to "installer" (cd installer)
Make few other scripts executable: chmod +x installer.* *.sh prl_*

OS X Screencast to animated GIF

This gist shows how to create a GIF screencast using only free OS X tools: QuickTime, ffmpeg, and gifsicle.

Instructions

To capture the video (filesize: 19MB), using the free "QuickTime Player" application:

Copy this file to: /Library/LaunchDaemons/fwd-80-to-3000.plist and then reboot:

$ sudo -s
$ curl -L https://gist.github.com/TooTallNate/3372589/raw/ace6451e9e47f59550f12d09cb924a64531cfd1f/fwd-80-to-3000.plist > /Library/LaunchDaemons/fwd-80-to-3000.plist
$ reboot

OS X Preferences

most of these require logout/restart to take effect

# Enable character repeat on keydown
defaults write -g ApplePressAndHoldEnabled -bool false

# Set a shorter Delay until key repeat

Bootstrap

Run software update
Download & install XCode in App Store
Install Google Chrome and enable syncing -- ensure 3rd party cookies are disabled
Install homebrew
brew install git
brew install wget
Copy ssh keys
git clone [email protected]:rmurphey/dotfiles.git

	#.bashrc

	alias disableGUI='sudo systemctl set-default multi-user.target'
	alias enableGUI='sudo systemctl set-default graphical.target'
	alias startGUI='sudo systemctl start gdm3.service'
	alias stopGUI='sudo systemctl stop gdm3.service'

	/usr/bin/setxkbmap -option "caps:escape"

	if [ -f ~/dev/personal/dot_files/.docker_aliases ]; then

	# losslessly concat mp4 files
	ffmpeg -i input1.mp4 -c copy -bsf:v h264_mp4toannexb -f mpegts intermediate1.ts
	ffmpeg -i input2.mp4 -c copy -bsf:v h264_mp4toannexb -f mpegts intermediate2.ts
	ffmpeg -i "concat:intermediate1.ts\|intermediate2.ts" -c copy -bsf:a aac_adtstoasc output.mp4

	# concat files with the same format
	# 1. create a file list
	file '/path/to/file1'
	file '/path/to/file2'
	file '/path/to/file3'

	import cv2
	import cv2.cv as cv


	def detect(img, cascade_fn='haarcascades/haarcascade_frontalface_alt.xml',
	scaleFactor=1.3, minNeighbors=4, minSize=(20, 20),
	flags=cv.CV_HAAR_SCALE_IMAGE):

	cascade = cv2.CascadeClassifier(cascade_fn)
	rects = cascade.detectMultiScale(img, scaleFactor=scaleFactor,