PixieDust 1.0 is here!

Notebook data visualizations for everyone

David Taieb
Mar 8, 2017 · 3 min read

Back in October of last year, I introduced the PixieDust open source project. PixieDust is a helper library for Python or Scala notebooks, which lets you generate sophisticated charts, maps, and other visualizations in a few clicks — no coding necessary. It smooths out some other pain points for notebook users too, which you’ll read about in a minute. PixieDust got lots of interest from the community. Thank you all for your feedback, which is helping us refine the tool.

The magic is bottled

Finally, the wait is over. After much hard work from the team, I am happy to announce the availability of PixieDust 1.0 on PyPi. Can’t wait to see it? Here’s a quick video showing PixieDust’s display() API and chart rendering in action:

New features

Multi-renderer support

PixieDust offers several different rendering engines you can use out-of-the-box to display your data. Depending upon what chart you’re viewing, render with matplotlib, Bokeh, or Seaborn — all without coding a single extra line. You can also generate sophisticated, gorgeous maps from your data using Mapbox or Google Maps.

Spark Progress Monitor

Track the status of your Spark job. No more waiting in the dark. Notebook users can now see how a cell’s code is running behind the scenes.

Installer for local use

We’ve made it easier to get started with PixieDust locally. Try our new packaged installer. It will walk you through setup, step by step.

Scala in a Python notebook

Enter Scala commands in a Python notebook. Variables are automatically transferred from Python to Scala and vice-versa.


Extensibility Guidance

Want to create your own visualizations or add a renderer? We help you understand how to build add-ons with a generate wizard, which walks you through a sample setup using Terminal or other command line tools.

Display improvements

We continue to refine and improve PixieDust’s display() API with smarter introspection of your DataFrames and expanded options for data visualizations.

That’s just the latest

As before, PixieDust lets you install Spark packages inside a Python notebook, export data, embed a polished app UI in your notebook, and more. For details on these and other features, visit PixieDust’s readme.

Coming soon

Full Scala notebook support

Love Scala, but crave the robust visualizations that only Python can deliver? Fear not, Matplotlib lovers — soon, there’ll be no need to choose!

PixieDust will soon work in Scala notebooks too, letting you configure robust and varied data display options in just a few clicks (no coding necessary). To see a preview, watch the video above or jump straight to it on YouTube.

Try it yourself

To help you get started, we offer some sample notebooks. Give PixieDust a try and share your issues, comments, and ideas on GitHub. PRs welcome!

Also, spread the magic. Click the ♡ here to sprinkle a bit of love in the name of PixieDust.

Oh, hey. Did we forget to mention? We have a new logo too:

Sprinkling data science magic since 2016.

Acknowledgements: Since there would be no magic without passion, I want to thank va barbosa, Mike Broberg, Jess Mantaro, Brad Noble, RAJ SINGH, Patrick Titzler, Chetna Warade, Mark Watson , and the rest of the of the IBM Watson Data Platform developer advocacy team for their dedication and long hours trying to make data simple and accessible.

Thanks to Brad Noble

David Taieb

Written by


Things we made with data at IBM’s Center for Open Source Data and AI Technologies.

Welcome to a place where words matter. On Medium, smart voices and original ideas take center stage - with no ads in sight. Watch
Follow all the topics you care about, and we’ll deliver the best stories for you to your homepage and inbox. Explore
Get unlimited access to the best stories on Medium — and support writers while you’re at it. Just $5/month. Upgrade