Awesome CMD Cheatsheets

Quick search Ctrl/Cmd + F the commands you need.

Linux

Check Linux version

lsb_release -a

Check Linux Architecture

uname -m

Change computer hostname

sudo nano /etc/hostname

Then update the old name with a new name.

sudo nano /etc/hosts

Then update any occurrence of the old name with a new name.

Reference

Check global disk storage usage

df -h

Check currunt disk storage usage

df -h .

Check all folder sizes recursively in the current folder

du -h

du -h .

Check folder sizes only in the current directory

du -h --max-depth=1 | sort -hr

Check top 20 largest directories under the current folder

du -Sh | sort -rh | head -20

Add a new path

export PATH=$PATH:<new_bin_path>

Example: <new_bin_path>=/home/user/.local/bin

export PATH=$PATH:/home/user/.local/bin

Display $PATH to verify

echo $PATH

Check CPU information

lscpu

Check hardware model

dmidecode | grep -A3 '^System Information'

Reference

Check RAM speed and type

sudo dmidecode --type 17

Reference

Check hard drive speed

sudo hdparm -Tt <Filesystem>

Example <Filesystem>=/dev/sdb2

sudo hdparm -Tt /dev/sdb2

Reference

Grant read, write and execute permissions for everyone, convenient for logging

chmod 777 <foldername/filename>

Grant read, write and execute permissions for everyone recursively, convenient for logging

chmod -R 777 <foldername/filename>

Enable syntax highlighting in vim on Mac

echo "syntax on" >> ~/.vimrc

Reference

Set default colorscheme in vim

Add colorscheme <colorscheme> in ~/.vimrc

Example <colorscheme>=industry

vim ~/.vimrc
colorscheme industry

Switch to colemak keyboard layout (1)

setxkbmap us -variant colemak

Reference

Switch to colemak keyboard layout (2)

cd ~/Downloads
wget https://colemak.com/pub/unix/colemak-1.0.tar.gz
tar -xf colemak-1.0.tar.gz
cd colemak-1.0
setxkbmap us; xmodmap xmodmap/xmodmap.colemak && xset r 66

Reference

Switch back to us-qwert keyboard layout

setxkbmap us

Swap Left Alt with Left Ctrl

sudo apt-get install gnome-tweak-tool
gnome-tweaks

Keyboard & Mouse -> Additional Layout Options -> Ctrl position -> Swap Left Alt with Left Ctrl

Reference

Reset function keys by default on Ubuntu using an Apple keyboard

echo 2 | sudo tee /sys/module/hid_apple/parameters/fnmode

Reference

Remap key

xev

Example: when pressing Left does not move left

KeyRelease event, serial 37, synthetic NO, window 0x5e00001,
    root 0x1e2, subw 0x0, time 96268495, (1670,972), root:(1904,1200),
    state 0x2010, keycode 113 (keysym 0xff7e, Mode_switch), same_screen YES,
    XLookupString gives 0 bytes: 
    XFilterEvent returns: False

xmodmap -e 'keycode 113 = Left'

Adjust mouse wheel scroll speed

sudo apt install imwheel
bash <(curl -s http://www.nicknorton.net/mousewheel.sh)

Reference

If the aforementioned url broke, try

sudo apt install imwheel
bash <(curl -s https://gist.hubusercontent.com/AshishKapoor/6f054e43578659b4525c47bf279099ba/raw/0b2ad8b67f02ebb01d99294b0ecb6feacc078f67/mousewheel.sh)

Reference

Stop laptop going to sleep when closing the lid

sudo nano /etc/systemd/logind.conf

Change

#HandleLidSwitch=suspend

HandleLidSwitch=ignore

then

sudo systemctl restart systemd-logind

Reference

Add a new user

sudo adduser <new_username>

Add a user to sudo group

usermod -aG sudo <username>

Switch to another user on one terminal

su - <another_user>

Install openssh server

sudo apt-get install openssh-server

Verify ssh service running

sudo systemctl status ssh

Enable ssh server

sudo systemctl enable ssh

Start ssh server

sudo systemctl start ssh

ssh via public/private key

also a alterniative solution to Permission Denied (publickey) problem.

Local Machine

mkdir ~/.ssh
chmod 700 ~/.ssh
cd ~/.ssh
ssh-keygen -t rsa

Terminal

Generating public/private rsa key pair.
Enter file in which to save the key (/home/<user>/.ssh/id_rsa): mykey
Enter passphrase (empty for no passphrase):
Enter same passphrase again:
Your identification has been saved in mykey.
Your public key has been saved in mykey.pub.

Enter key name (e.g. mykey on Terminal). Remember the passphrase yourself.

Executing the following command may still encounter the same issue of Permission denied (publickey).

ssh-copy-id -i ~/.ssh/mykey <user>@<remote_host>

Remote Machine

cd ~/.ssh
scp <user>@<local_host>:/home/<user>/.ssh/mykey.pub ~/.ssh/authorized_keys

Local Machine

ssh -i ~/.ssh/mykey <user>@<remote_host>

via passphrase

Reference

Count Files

ls | wc -l

Count Files in <foldername>

ls <foldername> | wc -l

List the first 5 files

ls -U | head -n 5

List the last 5 files

ls -U | tail -n 5

Reference: ChatGPT 3.5

Rename images as preparation for video making rename4video.py

Zip all folders in a directory and store the content in the corresponding folders with the same folder names

Create a zip_folders.sh:

#!/bin/bash

# Assuming your folders are in the current directory
for folder in */; do
    # Trim trailing slash to get folder name
    folder_name=${folder%/}

    # Create a zip file for each folder
    zip -r "${folder_name}.zip" "$folder_name"
done

bash zip_folders.sh

Reference: ChatGPT 3.5

Unzip all zipped files in a directory and store the content with the same folder name

find . -name '*.zip' -exec sh -c 'unzip -d "${1%.*}" "$1"' _ {} \;

[Reference]

Untar all tar files in a directory

for f in *.tar; do tar xf "$f"; done

[Reference]

Create a video from images

ffmpeg -framerate <FRAME_RATE> -i <DATASET_DEST_IMG_PATH>/image-%05d.png -c:v libx264 -profile:v high -crf 20 -pix_fmt yuv420p <DATASET_DEST_IMG_PATH>/RGB_video.mp4

Example FRAME_RATE=30 DATASET_DEST_IMG_PATH=/home/username/Data/RGB_copy/

ffmpeg -framerate 30 -i /home/username/Data/RGB_copy/image-%05d.png -c:v libx264 -profile:v high -crf 20 -pix_fmt yuv420p /home/username/Data/RGB_copy/RGB_video.mp4

List all disk devices

lsblk

e.g.

NAME   FSTYPE   LABEL  UUID                                 FSAVAIL FSUSE% MOUNTPOINT
loop0  squashfs                                                   0   100% /snap/bare/5
loop1  squashfs                                                   0   100% /snap/core18/1705
loop2  squashfs                                                   0   100% /snap/snap-store/433
loop3  squashfs                                                   0   100% /snap/snapd/7264
loop4  squashfs                                                   0   100% /snap/gnome-3-34-1804/24
loop5  squashfs                                                   0   100% /snap/gtk-common-themes/1506
loop6  squashfs                                                   0   100% /snap/gtk-common-themes/1535
loop7  squashfs                                                   0   100% /snap/core18/2745
loop8  squashfs                                                   0   100% /snap/gnome-3-34-1804/93
sda
├─sda1 vfat            55D0-8CD7                             504.9M     1% /boot/efi
└─sda2 ext4            53be03f3-b5bf-42a6-8f0c-0252d025f6b7  848.5G     2% /
sdb
├─sdb1 vfat            FACF-6821
├─sdb2
├─sdb3 ntfs            7C0A175A0A17112E
├─sdb4 ntfs            5CAC7AEAAC7ABDD8
├─sdb5 ntfs     新加卷 141A38A91A3889AE
└─sdb6 ntfs     新加卷 A2042BC1042B9777

Format Disk Partition /dev/sdb with ext4 File System

sudo mkfs -f ext4 /dev/sdb

Mount a new disk /dev/sdb1 not shown by df -h but listed by lsblk, ls /dev/sd* or sudo fdisk -l

sudo mkdir /ssh

Otherwise, it may encounter the issue of mount: /ssh: mount point does not exist. Then

sudo vim /etc/fstab

add the below line to the end of the file

/dev/sdb1 /ssd  ext4  defaults  0 0

sudo mount /ssd

Reference

If the following problem occurs

mount: /ssd: wrong fs type, bad option, bad superblock on /dev/sdb1, missing codepage or helper program, or other error.

do (note that mkfs.ext4 will erase data!)

mkfs.ext4 /dev/sdb1

and sudo mount /ssd again, you will see the new disk by df -h

Reference

If ext4 does not work, try exfat format

Reference

Rename the directory name listed under the Mounted on column after df -h

Example: remove _ from /_eDrive where /_eDrive is the current directory of the external drive to be renamed.

df -h

Filesystem      Size  Used Avail Use% Mounted on
...
/dev/sdc        1.8T  1.5T  306G  83% /_eDrive

sudo umount /_eDrive
sudo mkdir /eDrive
sudo vi /etc/fstab

Append

/dev/sdc /eDrive        ext4    defaults        0       0

to the end of this file.

sudo mount /dev/sdc

Then run df -h and you should be able to see

Filesystem      Size  Used Avail Use% Mounted on
...
/dev/sdc        1.8T  1.5T  306G  83% /eDrive

Reference

Kill all background jobs

kill $(jobs -p)

Reference

Download files from Google Drive

pip install gdown

gdown https://drive.google.com/uc?id=<file_id>

Example https://drive.google.com/file/d/1h_8Ts11rf0GQ4_n6FgmCeBuFcWrRjJfa/view

gdown https://drive.google.com/uc?id=1h_8Ts11rf0GQ4_n6FgmCeBuFcWrRjJfa

Reference

Setup Razer Core X

pcie_ports=native pci=assign-busses,nocrs,realloc iommu=on
sudo update-grub
sudo add-apt-repository ppa:hertg/egpu-switcher
sudo apt install egpu-switcher
sudo egpu-switcher setup

Then Reboot the system.

Reference

Install Docker on Ubuntu 18.04

Version Checking

Check CUDA version

nvcc --version

nvcc -V

ls -l /usr/local | grep cuda

Check CUDA version - Prerequisite

sudo apt install nvidia-cuda-toolkit

Check TensorFlow version

pip list | grep tensorflow

python -c 'import tensorflow as tf; print(tf.__version__)'

Check Keras version

pip list | grep Keras

python -c 'import keras; print(keras.__version__)'

Check PyTorch version

python -c 'import torch; print(torch.__version__)'

Check OpenCV version

python -c 'import cv2; print(cv2.__version__)'

Git

Set global user email

git config --global user.email "<you@example.com>"

Set global username

git config --global user.name "<Your Name>"

Save user's credential

git config credential.helper store

Conda

List env names

conda info --envs

Clone env <env_name> from machine source to destination

In source machine

conda list --explicit > <env_name>.txt

Example <env_name>=cln_env

conda list --explicit > cln_env.txt

Then cln_env.txt to destination machine

In destination machine

conda create --name <env_name> --file <path_to_><env_name>.txt

Example <env_name>=cln_env under the <path_to_> folder

conda create --name cln_env --file cln_env.txt

Using Jupyter Notebook on Local/Remote Machines

Launch jupyter notebook kernel on a remote server

jupyter notebook --no-browser --port=<PORT_NUM>

Example <PORT_NUM>=8888

jupyter notebook --no-browser --port=8888

Launch jupyter notebook kernel with a specific GPU on a remote server

CUDA_VISIBLE_DEVICES=<GPU_NUM> jupyter notebook --no-browser --port=<PORT_NUM>

Example <GPU_NUM>=0 <PORT_NUM>=8888

CUDA_VISIBLE_DEVICES=0 jupyter notebook --no-browser --port=8888

ssh on a local machine

ssh -N -f -L 127.0.0.1:<PORT_NUM>:127.0.0.1:<PORT_NUM> <USER>@<URL>

Example <PORT_NUM>=8888 <USER>=admin <URL>=192.168.1.3

ssh -N -f -L 127.0.0.1:8888:127.0.0.1:8888 admin@192.168.1.3

Then open a browser on the local machine with the URL provided on the Terminal, e.g. *localhost:8888*

Kill port in use

lsof -ti:<PORT_NUM>

Example <PORT_NUM>=8880

lsof -ti:8880

Training with GPUs

Train a model with a specific GPU

CUDA_VISIBLE_DEVICES=<GPU_NUM> python train.py

Example <GPU_NUM>=0

CUDA_VISIBLE_DEVICES=0 python train.py

Check Running Containers

docker ps -a

(base) <user>@<machine>:~$ docker ps -a
CONTAINER ID   IMAGE          COMMAND          CREATED         STATUS                       PORTS     NAMES
<CONTAINER_ID>   <IMAGE_ID>   "/bin/bash"      6 minutes ago   Up 6 minutes                           funny_mcnulty

Run a Container with Access to GPUs

docker run --ipc=host --shm-size=16384m -it -v /:/share --gpus all --network=bridge <IMAGE_ID> /bin/bash

nvidia-docker run --ipc=host --shm-size=16384m -it -v /:/share --network=bridge <IMAGE_ID> /bin/bash

However, when you exit the container by exit, it will also shut down.

Run a Container with Access to GPUs in Detached Mode `-d`

If you need to regularly exit and enter a container, it is recommended to run it in detached mode (in the background) with -d:

docker run -d --ipc=host --shm-size=16384m -it -v /:/share --gpus all --network=bridge <IMAGE_ID> /bin/bash

nvidia-docker run -d --ipc=host --shm-size=16384m -it -v /:/share --network=bridge <IMAGE_ID> /bin/bash

Bash into a Running Container

docker exec -it <CONTAINER_ID> /bin/bash

Using TensorBoard Remotely

Run Tensorboard on a remote server

tensorboard --logdir=runs

Then check the displayed on the Terminal.

ssh on a local machine:

ssh -N -f -L 127.0.0.1:<PORT_NUM>:127.0.0.1:<PORT_NUM> <USER>@<URL>

Example <PORT_NUM>=6006 <USER>=admin <URL>=192.168.1.3

ssh -N -f -L 127.0.0.1:6006:127.0.0.1:6006 admin@192.168.1.3

Then open a browser on the local machine with the URL provided on the Terminal, e.g. *localhost:6006*.

Atom

Replace delete-line shortcut ctrl-shift-k with ctrl-d

Edit -> Preferences -> Keybindings

Search ctrl-shift-k in the "Search keybindings" box, find the command of delete-line

editor:delete-line

Search ctrl-d (which you want to use) in the "Search keybindings" box

Click your keymap file, then it will navigate to .atom/keymap.cson, enter the following text in this file

Linux:

'atom-text-editor:not([mini])':
  'ctrl-d': 'editor:delete-line'

macOS

'atom-text-editor:not([mini])':
  'cmd-d': 'editor:delete-line'

Save the file .atom/keymap.cson and restart Atom to take effect.

Select multiple lines

Cmd + Shift + L

Combine multiple lines in one row

Cmd + J

remote-atom

local machine

sudo apm install remote-atom

Open Atom app: Packages -> Remote Atom -> Start Server

remote server

sudo curl -o /usr/local/bin/rmate https://raw.githubusercontent.com/aurora/rmate/master/rmate
sudo chmod +x /usr/local/bin/rmate

Usage

local machine

ssh -R 52698:localhost:52698 <user@example.com>

remote machine

rmate <test.txt>

Reference

Miscellaneous

Write latex-like math equations in Markdown in Github *.md files

<img src="https://render.githubusercontent.com/render/math?math={<latex-like_equations>}">

Fill the placeholder <latex-like_equations>. Example

<img src="https://render.githubusercontent.com/render/math?math={\kappa_{\Theta}=\frac{\lambda_{max}}{\lambda_{min}}}">

Display

${\kappa_{\Theta}=\frac{\lambda_{max}}{\lambda_{min}}}$

Reference

Make hard disk compatible (read & write) in both Mac or Linux (Ubuntu) systems: choose partition format as ExFAT when erasing the whole disk on Mac.

bryanbocao / awesome-cmd-cheatsheets

readme

Awesome CMD Cheatsheets

Linux

Version Checking

Git

Conda

Using Jupyter Notebook on Local/Remote Machines

Training with GPUs

Check Running Containers

Run a Container with Access to GPUs

Run a Container with Access to GPUs in Detached Mode `-d`

Bash into a Running Container

Using TensorBoard Remotely

Atom

remote-atom

Usage

Miscellaneous

bryanbocao / awesome-cmd-cheatsheets

readme

Awesome CMD Cheatsheets

Linux

Version Checking

Git

Conda

Using Jupyter Notebook on Local/Remote Machines

Training with GPUs

Check Running Containers

Run a Container with Access to GPUs

Run a Container with Access to GPUs in Detached Mode -d

Bash into a Running Container

Using TensorBoard Remotely

Atom

remote-atom

Usage

Miscellaneous

Run a Container with Access to GPUs in Detached Mode `-d`