Skip to content
Tom Dupré la Tour edited this page Jun 29, 2017 · 24 revisions

Welcome to Telecom

This is a list of things you should know it exists.

TPT administration

  • Eole: TPT intranet (in french)
  • Zimbra: TPT webmail
  • Min&Tel: another TPT intranet (in french)
    • This is where you can declare your holidays.
  • TPT: TPT public website
    • You can find some informations about accomodation, healthcare, cost of living, ...
  • IDS: IDS department intranet (in french)

TPT account

  • DSI: You can ask any question about your account, your desktop, ...
  • Personal website: The DSI can host your personal website at https://perso.telecom-paristech.fr/your_login
    • GitHub can also host your personal website at https://your_github_login.github.io/
  • Home: Your home directory is accessible from most machines in TPT, but its size is limited to ~8 Gbytes
  • More space: You can ask the DSI for more space on your local desktop, or on a shared directory in /tsi/doctorants/your_login
  • Wifi connection: To setup wifi in TPT (in french)
    • You can also use your email address (name.surname@telecom-paristech.fr) and your TPT password to connect to the eduroam wifi, available in most french universities.

TPT ssh and clusters

  • SSH:
    • To access TPT by SSH from outside TPT, use ssh your_login@ssh.enst.fr; you are now "inside".
    • To access a specific machine, e.g. lame10, use ssh your_login@lame10. Note that you need to be "inside" TPT.
    • To do both step in one, you can easily redirect your ssh (Linux).
  • Infres cluster: List of machines you can access by ssh, without any queuing system.
    • lame10-16: You have your home folder unchanged, so it's very easy to use.
    • lame17-19: You need to install what you need, your home is not accessible.
    • lame20-22: You can book these machines on this wesite.
  • IDS cluster: A cluster you can access by ssh, but you need to use the queue
    • Please read the doc before using it ! (in french)

Working tools

Python tools

  • Anaconda: A good way to have everything you need for python
    • Miniconda: To have the conda package manager, but not all the default packages of Anaconda
  • Windows binaries: Unofficial packages compiled for Windows.
  • Scientific basic packages: ipython, numpy, scipy, pandas, jupyter, matplotlib
  • Editing tools: pep8, flake8, yapf
  • Profiling: snakeviz, line-profiler, memory-profiler
  • Editors: sublime_text/atom, spyder/pycharm (Matlab style)
  • Awesomeness: list of links

Clone this wiki locally