Introduction to Jupyter

Goals:

• Introduce what Jupyter is and why it’s useful

What is Jupyter?

Jupyter, formerly known as an ipython notebook, is a popular tool used in data science and data analysis.

• An open-source, browser-based, web application with a wide variety of functions

  • Allows users to create and share computational documents, called notebooks.

  • Notebooks facilitate the development of live code that can then be run in a number of different coding languages.

    • Code can be run step by step in “chunks” called cells.

  • Users can combine live code cells with other cells - Markdown text, images, plots, and other rich media in a single interactive canvas.

  • Can produce a wide variety of interactive output including HTML, videos, LaTeX, and custom MIME types.

  • Can be shared through e-mail, GitHub, or other cloud storage and sharing services.

  • Easily exported to other formats like, books, slides, web apps, static web pages, or PDF documents.

• We see this tool used for a number of things:

  • To organize work and display the thought process or logic associated with a project

  • Collaboration

  • As an IDE (Integrated Development Environment)

    • Easy to look at or run code line-by-line to simplify debugging

  • For teaching

  • Displaying and manipulating data frames

Jupyter Notebooks

• Jupyter Notebooks are just the Notebooks themselves (which on their own are great), but lack some of the functionality of Jupyter Lab

  • Simple interface where users can open and run notebooks, terminals, and text files

  • A web based interactive environment in which to create Jupyter notebook documents

  • Launched from a kernel

Jupyter Labs

• Jupyter’s “next generation interface.

• Includes notebooks, but extends to consoles, terminals, CSV editors, markdown editors, interactive maps, and more.

• Has a modular structure, allowing you to open several notebooks and added files like HTML, Text, markdown in the same window - more like an IDE.  

• Lab also allows users to execute code in a python console 

Jupyter Notebook cell types

• Markdown

  • Text Cells allowing you to write and render Markdown syntax

  • This is where you describe and document your workflow

• Code

  • Code cells allow you to write and run programming code in a language of your choosing (e.g., Python)

  • Languages supported in Jupyter include Python, R, Julia, and many others

  • On ARCC HPC resources, we support jupyter code in Python and R

• NBConvert

  • Stands for Notebook Convert

  • Enables the conversion of your notebook to another format as given by the FORMAT string using Jinja templates.

    • Presenting: PDF

    • Publishing: LaTeX

    • Collaboration

    • Sharing: HTML

Running Jupyter from OnDemand

After logging into OnDemand on your favorite ARCC HPC resource, you can launch Jupyter from the main Dashboard:

Upon request, you are taken to a web form to tailor and specify the Jupyter environment you’d like to run in your session