universome

#!/bin/bash
forbidden_extensions=( "jpg" "jpeg" "png" "gif" "swf" "svg" )

# Usage: includes 'blabla' $arr
includes () {
  local e
  for e in "${@:2}"; do [[ "$e" == "$1" ]] && return 0; done
  return 1
}

universome

$(".lessons-left")
  .map((i,el)  => $(el).text())
  .toArray()
  .filter(Boolean)
  .map(s => s.split('/')[1])
  .map(Number)
  .reduce((a,b) => a + b);

universome

sed -n '1!p' data.csv | perl -MList::Util=shuffle -e 'print shuffle(<STDIN>);' > shuffled.csv
head -1 data.csv >> tmp & cat shuffled.csv >> tmp
mv tmp shuffled.csv
# You can also use "awk 'NR>1' data.csv", but it will be slower

universome

###
# Stop and remove all containers
docker stop $(docker ps -a -q) && docker rm $(docker ps -a -q)
###

###
# Run docker daemon (if you see stupid error "Cannot connect to the Docker daemon. Is the docker daemon running on this host?")
# Ubuntu 14.04:
sudo service docker start

universome

sudo apt-get update
sudo apt-get install python3-pip python-virtualenv virtualenv

# and now you can use it via:
# virtualenv -p python3 kek

universome

sudo apt-get update
sudo apt-get -y install python3-pip python-virtualenv python-dev

mkdir virtualenvs
virtualenv -p python3 virtualenvs/kek

git clone git@github.com:kek/kek-repo.git kek
cd kek
source ~/virtualenvs/kek/bin/activate

universome

# I dunno why, but we should specify 0.0.0.0 IP address for jupyter
# when we want to login remotely (for example, aws ec2 or docker)
jupyter-lab --ip=0.0.0.0 --no-browser --port=10000

# In case HTTP ports are not open in the server —
# use ssh forwarding on local machine
ssh -N -f -L localhost:8888:localhost:10000 user@host

universome

# Creating pyvenv
python3.6 -m venv --without-pip ~/pyvenvs/zoo
source ~/pyvenvs/zoo/bin/activate

# Getting pip inside
curl https://bootstrap.pypa.io/get-pip.py | python

# Reactivating to obtain pip from venv (without it we would use global pip)
deactivate && source ~/pyvenvs/zoo/bin/activate

universome

from torch.distributions import MultivariateNormal
from torch.distributions import kl_divergence

mean = torch.rand(10, 100)
logstds = torch.rand(10, 100)
x = torch.rand(10, 100)

dist = MultivariateNormal(mean, scale_tril=torch.stack([torch.diag((ls).exp()) for ls in logstds]))
print('Testing LL')
print(dist.log_prob(x).mean())

universome

{
    "shortcuts": [
        {
            "command": "documentsearch:start",
            "keys": [
                "Accel F"
            ],
            "selector": ".jp-mod-searchable",
            "disabled": true
        }