Jupyter Notebook for Data Science Teams

April 28th, 2016

I just released a new screencast course for O’Reilly Media: Jupyter Notebook for Data Science Teams!

First, some background: the Jupyter Notebook (evolved from the IPython Notebook) has been a favorite tool of people who use Python, R, Julia, and many of the other languages that it supports. Data scientists and researchers, in particular, have taken up Notebooks. There are many reasons it’s so popular, but to pick a couple:

  • The inline plotting that shows the output in the same document that one codes in allows rapid data visualization and quick iteration of code ideas.
  • You can export the Notebook to many formats, and can be easily shared with people as a single HTML file that’s readable by any web browser.

It’s one of my favorite tools for the job, and I’ve spent the past few years working with the Jupyter Notebook daily by myself and in data science teams at SVDS. We explored, analyzed, and iterated on visualizations Caltrain data with the Jupyter Notebook, and we’ve previous posted a recommended Jupyter Notebook workflow.

Fully grokking the Jupyter Notebook takes concerted effort and practice. The Jupyter Notebook for Data Science Teams screencast is aimed at an intermediate level audience, but will also be useful to people who consider themselves advanced. The course is just over three hours of video, and it covers tips and tricks to gain efficiency in everyday data science use. Finally, there is also special consideration on how to organize use of the Jupyter Notebook when multiple Data Scientists are working together as a team.

Let us know how you’re using Notebooks, either in the comments here or by getting in touch.