Coding in python in a Jupyter Notebook

• I made some slides showing how to get started
• check out the jupyter website for docs and the like
• for further info on the jupyter extensions check out their github page
• feel free to download and play around with the notebook we worked on; (tarball contains both the notebook and the fits file)

Installation

• assuming you have 'pip' installed run the following

pip jupyter install

• if you'd like to use the extensions we discussed you'll also need to

pip install jupyter_contrib_nbextensions
jupyter contrib nbextension install --user

• optionally you can also create a config file like so

jupyter notebook --generate-config

• this is useful in case you want to run jupyter not in your default web browers; to get this to work you'll need to modify the file $HOME/.jupyter/jupyter_notebook_config.py by changing the follwing line according to your needs: c.NotebookApp.browser = '/usr/bin/browser'
• if you'd like to make use of widgets like ipywidget you'll need to have them install and enable them in jupyter:

pip install ipywidgets
jupyter nbextension enable --py widgetsnbextension

start/stop a notebook

• from a console run jupyter notebook
• this will open up a tab in your favourite web browser, check out the pdf on how to continue from here
• to stop the jupyter server you have to hit Ctrl-c

Run jupyter on remote machine

• let's assume you need to run/test code on a remote machine and you want to do this in jupyter notebook
• what you'll need to do is start the jupyter server like this: jupyter notebook --port=5558 --ip=127.0.0.1
• this will send the X-window to localhost on port 5558 (note, it does not need to be on port 5558)
• on your local machine you create a SSH tunnel mapping the remote X-session onto some local port: ssh -L 5558:localhost:5558 user@remoteMachine
• you open the notebook in your browser of choice using the following as the URL: http://localhost:5558/tree#