For windows users: Windows equivalent to Linux commands
copying
scp -rCq Desktop/cs231n rakhil@xxx.xxx.xxx.xxx:Desktop/
rsync -avz -e 'ssh -i x.pem' Desktop/cs231n rakhil@xxx.xxx.xxx.xxx:Desktop/
rsync -avz Desktop/cs231n rakhil@xxx.xxx.xxx.xxx:Desktop/
#use rsync if you want to update files in the same directory
#if you want delete old files in the same directory use --delete
#add --rsync-path="sudo rsync" if ur copying to sudo location
#faster rsync
rsync -aqzP C3D rakhil@xxx.xxx.xxx.xxx:~/Desktop/
tar
tar -xvzf dasd //extract
tar -cvzf dasd //compress
ssh using pem file
#or use aliases in /etc/hosts and .ssh/config
ssh -i xxx.pem rakhil@xxx.xxx.xxx.xxx
curr time
date
root mode
sudo su
appending file
echo 'dadaf3dad' >> lol.txt
history of commands
history | grep "string"
vim ~/.bash_history
#reverse search with word
ctrl + r ; type string
files n space in directory
du -hs #space in directory
find . -type f | wc -l #num files in directory
df -h #available space in disk
list
ls
ll #advanced list
ll -lrt #sort with time
Check quota alloted to user
quota -us
ip
curl ipinfo.io/ip #global ip
ifconfig #get local ip
ipconfig #local ip forwindows
going to D drive in windows.. no cd
D:
vim
cursor to start: gg; end: G; end of line: $
del line: dd
del all: dG (from cursor to end)
search: /search_word ; n & shift + n or N for next and prev word
diff between files
#use meld in ubuntu or
vimdiff lol1.txt lol2.txt
for viewing images
eog lol.jpg
check symbolic links
ls -la /usr/bin/ | grep gcc
ll
creating symbolic links
ln -sfn /usr/bin/gcc-4.3 /usr/bin/gcc
env variables
printenv #list all
echo $LD_LIBRARY_PATH #print
export LD_LIBRARY_PATH=$LD_LIBRARY_PATH:/rak/lib #add new path
browser from terminal
sensible-browser
wget
echo "check_certificate = off" >> ~/.wgetrc
youtube-dl
youtube-dl -f best -f mp4 https://www.youtube.com/watch?v=ekyBklxwQMU --proxy xxx.xxx.xxx.xx:80 --no-check-certificate
# mp3 format
youtube-dl -x --audio-format mp3 -i --yes-playlist https://www.youtube.com/playlist?list=PLNPDZJ1Lllc9bIW77JV9JEotb7_FR40pZ
youtube-dl -x --audio-format mp3 -i PLNPDZJ1Lllc9bIW77JV9JEotb7_FR40pZ
in sh file
#!/bin/bash #for bash script
#!/bin/sh #for shell script
#executing bash files
sh lol.sh
#executing shell script
chmod +x lol.sh
./lol.sh
reading input from txt file in cpp
freopen("input.txt", "r", stdin);
adding a user in ubuntu
sudo su
adduser username
usermod -aG sudo username #give sudo access
su - username #switch to new user
sudo ls -la /root #check if sudo access given
search for string in all files
grep -r "lol"
grep -r "lol" . #in curr dir
grep -i "lol" #ignore-case
#search file
locate lol.txt
accessing n downloading files through browser
python -m SimpleHTTPServer 8080 #in remote server
xxx.xxx.xxx.xxx:8080 #open in local browser
python libraries location
/usr/local/lib/python2.7/
pip install locally
export PIP_CERT=/data/CRT_New.cer #if cert required
pip install lol --user
python -m pip install --user --upgrade lol
using local library without installation
import sys
sys.path.insert(0,'/data/rakhil/keras')
using linux commands
os.system('mkdir lol')
os.makedirs(lol, exist_ok=True)
Try and expect
import traceback
try:
# some error occurs
except Exception as e:
print(traceback.format_exc())
mkvirtualenv lol
workon lol
deactivate
rmvirtualenv lol
ls $WORKON_HOME
pull
git status
git remote update
git pull
push
git add -u #add only edited files
git add new_file.py #add a new specific file
git add . #add all files
git commit -m "lol"
git push -u origin master
Git to overwrite local files
git fetch origin master
git reset —hard FETCH_HEAD
git clean -df
Git pull only one file
git fetch
git checkout FETCH_HEAD lol.py
tmux new -s name #create new tmux session
tmux attach -t name #attach a specific tmux session.. need not type full name, unique prefix is enough
tmux kill-session #in session or ctrl+d
detach: ctrl+b d
mouse up/down: ctrl+b [
next window: ctrl+b n prev: ctrl+b p
new window: ctrl+b c kill window: ctrl+b x
show tmux sessions in tmux: ctrl+b s
gpu info
lspci | grep VGA
using gpu
#in terminal
echo $CUDA_VISIBLE_DEVICES
export CUDA_VISIBLE_DEVICES=1,2
#in python script
os.environ["CUDA_VISIBLE_DEVICES"] = "3"
gpu usage with refresh interval 1s
nvidia-smi -l 1
without filling terminal
watch -n 1 nvidia-smi
user process running on GPU
ps -f `nvidia-smi | grep -A 100 PID | grep C | awk {'printf($3"\t")'}`
cuda and cudnn versions
nvcc --version #cuda version
cat /usr/local/cuda/include/cudnn.h | grep CUDNN_MAJOR -A 2 #cudnn version
ipython in remote machine and using it from local machine;go to browser localhost:8888
remote_user@remote_host$ ipython notebook --no-browser --port=8890
local_user@local_host$ ssh -N -L localhost:8888:localhost:8890 remote_user@xxx.xxx.xxx.xxx
Common access system
ipython notebook --ip '*' --port 8890 --no-browser
auto reload ipython
%load_ext autoreload
%autoreload 2
for using system commands in ipython
!echo lol
kill jupyter kernel from inside the notebook
%%javascript
Jupyter.notebook.session.delete();
vim ~/.keras/keras.json
THEANO_FLAGS=device=gpu0,lib.cnmem=0.55 python keras_proposal_layer.py #theano
CUDA_VISIBLE_DEVICES=3 python train_rpn.py #tensorflow
#caffe suppress output prototxt https://stackoverflow.com/questions/29788075/setting-glog-minloglevel-1-to-prevent-output-in-shell-from-caffe
#use before import caffe
os.environ['GLOG_minloglevel'] = '2'
gpu memory control
config = tf.ConfigProto()
config.gpu_options.allow_growth = True #takes the min required memory or
config.gpu_options.per_process_gpu_memory_fraction = 0.4 #fixed memory - 40% of total memory
session = tf.Session(config=config, ...)
test gpu
tf.test.is_gpu_available()
saving and restoring tf model: http://cv-tricks.com/tensorflow-tutorial/save-restore-tensorflow-models-quick-complete-tutorial/
tensorboard port forwarding on remote machine(6006) and viewing on local machine port(16006)
ssh -f -L localhost:16006:localhost:6006 <user@remote> #in local
ps aux | grep 6006 #for closing connection #in remote
kill -9 pid #in remote
octave -q test.m
or in octave command line:
test
installing locally
./configure --prefix=/home/maxpower/.octave38/
make -j2
make install
echo "alias octave38='~/.octave38/bin/octave'" >> ~/.bashrc
uninstall
make uninstall
remove alias and downloaded folder
ffmpeg -i lolo.mp4 #print properties
ffmpeg -i inp.mp4 -ss 00:30:00 -t 00:50:00 -acodec copy -vcodec copy out.mp4 //cut video of 50 secs starting from 30s
ffmpeg -i 1.mp4 -i 2.mp4 -i 3.mp4 -filter_complex "[0:v] [0:a] [1:v] [1:a] [2:v] [2:a] concat=n=3:v=1:a=1 [v] [a]" -map "[v]" -map "[a]" output_3.mp4 #combine diff videos
ffmpeg -f concat -i mylist.txt -c copy output.mp4 #in list.txt format: file 'path/to/file'; combine similar videos
make socket connection
wscat -n -c wss://PRD_XXX.elb.amazonaws.com:443
socket processes
netstat -tulpn #or
netstat -anp
ps -ef | grep 3000
sudo netstat -anp | grep :3000 | grep ESTABLISHED | wc -l
forever list
systemctl status nginx
systemctl start nginx
sudo cat /var/log/syslog | grep nginx #check logs
enable service
sudo chmod 755 /etc/init.d/nginx
sudo systemctl enable nginx
sudo systemctl daemon-reload
sudo systemctl reset-failed
install sublime text 3 without sudo on ubuntu server https://forum.sublimetext.com/t/install-sublime-text3-without-sudo/42893/4