Using Jupyter at Jefferson Lab (OUTDATED)

This Sowftware Carpentry Tutorial is based on the state of the Jefferson Lab Jupyter service in 2018. The instructions below are NOT guaranteed to be valid, but remain accessible here for archival purposes. Please consider the official documentation at https://scicomp.jlab.org/docs/JupyterHub when starting to use the Jefferson Lab Jupyter service.

Jupyter Notebooks present a powerful way to run Python code interactively from anywhere in a web browser environment. Because of the large size of data files in use at Jefferson Lab, there are significant benefits to using Jupyter Notebooks on jupyter.jlab.org to access data files on the large file data storage systems directly.

At the end of this lesson you will be able to:

1. Create new Jupyter Notebooks on Jefferson Lab’s Jupyter server.
2. Increase functionality by installing additional Python packages.
3. Use Git for version control of Jupyter Notebooks.

Prerequisites

Learners need to understand the concepts of files and directories (including the working directory), have a basic understanding of the Python programming language, and of the Jupyter (IPython) Notbeook interface before tackling this lesson. The commands in this lesson pertain to Python 3.

Learners should ensure that they are part of the jupyterusers unix group on the Jefferson Lab scientific computing system. Use the groups command in an interactive session to check if you are a member of this group.

Getting Started

To get started, follow the directions in the “Setup” tab to ensure that you are in the jupyterusers unix group.

In order to speed up the entry of notebooks, later in the lesson we will download an initial set of Jupyter notebooks to your working directory on the server.