PUN Meetup - Amsterdam 13 Nov 2013

For the past eight months I have been using pandas on and off for analysing the results of questionnaires. Pandas has left such a great impression on me that I felt compelled to introduce more Python developers to this wonderful library. The Python Users Netherlands (PUN) meetup in Amsterdam on November 13th, 2013 provided that opportunity.

Although I use PyCharm for pretty much all my Python development, I have found that IPython Notebook provides the best environment for writing my pandas code. Due to its expressiveness pandas code tends to do a whole lot in relatively few lines of code. At times it can be hard to visualize the current state or form your data structures are in. IPython Notebook greatly helps by allowing you inspect your data structures by printing them out or plotting them. It allows you to change parts of your code, re-execute only those parts, and immediately see the results. This is what it must have been for all those Lisp and Smalltalk developers that always rave about their image based integrated development environments.

Last week I found out IPython Notebooks can even be converted to a reveal.js based slideshows. That meant I could write the code for my talk in the same environment/document that would become my presentation. It seemed to work well and all was well on my laptop. Though when I connected my laptop to the projector the fonts were too small for the crowd in the back of room. For some reason CMD-+ in Firefox only marginally zoomed in and I had to resort to using the OS X built in zoom feature (CTRL + scroll up/down). Zooming in and out on every slide must have been an erratic viewing experience for the attendees; my apologies.

I have slightly adjusted the IPython Notebook that I used for my presentation by removing all the slideshow metadata/markers. Removed some of the headers now that it is one big continuous document without specific slides. And added a comment here and there, though probably not enough to be all self-explanatory:

Data Analysis with pandas

Comments !