[IPython-dev] A fresh approach for plotting?

Shead, Timothy tshead at sandia.gov
Wed Apr 16 16:22:22 EDT 2014


I am very new to IPython, but it’s had a profound impact on how I do my work, in a very short time.  Lately, I’ve been generating lots of animated plots using Matplotlib, and JSAnimation to embed them in my notebooks.  This works great, but the way JSAnimation does it (it embeds each frame of the matplotlib animation as a base64-encoded image into the HTML)  leads to excessively large notebooks - my animations typically weigh in around 40MB with this approach.  BTW, this is not a dig at Jake Vanderplas, who wrote JSAnimation - when I realized how he had integrated it into the IPython rich display, I was blown away by the possibilities :-)

So anyway, I’ve started working on a toy plotting library that generates SVG plots with animation, and integrates well with notebooks.  You can see and play-back a live demo at http://nbviewer.ipython.org/gist/anonymous/b7a626c3f6cac6d2be3b 

Even though it’s in the very early stages, I think that letting-go of bitmaps and targeting browser + SVG as the primary backend is an interesting new direction for a plotting library.  The results are lightweight, gorgeous on a retina display, and introduce new possibilities for interaction.  I’ll admit, it’s also a guilty pleasure to design my own API and get rid of the Matplotlib cruft ;-)  I’d be curious to get some opinions on whether this idea could have traction.  Is anyone else taking this approach? 

Many thanks,

Timothy M. Shead
Sandia National Laboratories
1461, Scalable Analysis and Visualization

-------------- next part --------------
A non-text attachment was scrubbed...
Name: smime.p7s
Type: application/pkcs7-signature
Size: 3216 bytes
Desc: not available
URL: <http://mail.python.org/pipermail/ipython-dev/attachments/20140416/6aa2f452/attachment.bin>

More information about the IPython-dev mailing list