jeremyfix

name: torchcomplex
channels:
  - pytorch
  - defaults
dependencies:
  - _libgcc_mutex=0.1=main
  - _openmp_mutex=5.1=1_gnu
  - blas=1.0=mkl
  - brotli=1.0.9=h5eee18b_7
  - brotli-bin=1.0.9=h5eee18b_7

jeremyfix

import pathlib
import PIL.Image as Image
import numpy as np
import matplotlib.pyplot as plt
from matplotlib import cm
import json


# Code taken from utils.py on https://github.com/alexsax/2D-3D-Semantics
def get_index(color):

jeremyfix

# coding: utf-8

# External imports
import numpy as np

FIELD_WIDTHS = {
    'int': 16
}
ENDIAN = 'big'

jeremyfix

# The margins were on the right and bottom parts (hence the first rotation to ease margin removal)

convert -rotate 180 -crop +130+210 +repage -density 130 -rotate -90 -colorspace gray +repage source.pdf tmp.pdf && mutool poster -x 2 tmp.pdf target.pdf

jeremyfix

# Make the workspace
mkdir ~/catkin_ws
cd catkin_ws
mkdir -p src
catkin_make
cd src

# Clone the required repositories
git clone https://github.com/turtlebot/turtlebot.git

jeremyfix

To run a jupyter lab with a specific token, just :

jupyter lab --NotebookApp.token='toto'

This is handfull to use a custom token , for example when running a batch of jupyter lab for students working remotely to which you just want to provide a token and not the quite complicated instance specific URL.

jeremyfix

Given texts from wikisource, the raw one , with the markup, from here https://fr.wikisource.org/w/index.php?title=Fables_et_opuscules_p%C3%A9dagogiques&action=edit

and suppose the whole wiki text is copied in fenelon.txt , we can somehow get rid of the markup and so on with the following

cat fenelon.txt | sed '/^</d' | perl -pe 's|\[\[Auteur.*?\]\]||' | sed 's/\[\[//g' | sed 's/\]\]//g' | sed 's/==//g' | sed '/^$/d' | perl -pe 's|<ref>.*?</ref>||' | perl -pe 's|<ref>.*?</ref>||' | perl -pe 's|<sup>.*?</sup>||' | perl -pe 's|</.*?>||' | sed "s/\’/'/g" | sed 's/»/"/g' | sed 's/«/"/g' | perl -pe 's|\{\{.*?\}\}||' | sed 's/œ/oe/g' | sed 's/Œ/OE/g' > fenelon2.txt

there are still some markup though.

jeremyfix

# Description of the issue :

After pushing a text file, which was UTF-8 encoded (as indicated by the file command) on github and pulling it back on another computer on which I was connected through ssh , 
I was unable to see the accents in the files (these appeared either as _ _ _ or <C5><A9>, etc..)

Then , I added to my bashrc 

export LANG="en_US.UTF-8"
export LC_ALL="en_US.UTF-8"

jeremyfix

# Extract, e.g., the second page
pdftk input-file.pdf cat 2 output pageA3.pdf

# Scale it up
pdfjam --outfile pageA4.pdf --paper a3paper pageA3.pd

# Split it in mutiple A4 pages 
pdfposter pageA3.pdf multipleA4.pdf -m a4 -p 2x1a4

jeremyfix

For copying your ssh keys to facilitate the configuration

for i in {1..59}; do sshpass -p THE_SECRET_PASSWORD ssh-copy-id myacount_$i@mymachine.mydomain.com ; done

For setting up the xstartup script

  for i in {1..59}; do ssh myacount_$i@mymachine.mydomain.com << 'EOF'