Really loving these ipython visualisation articles!

There was one about detecting bubbles in liquid for the purposes of reducing spilling that was very interesting

http://nbviewer.ipython.org/github/soft-matter/trackpy-examp...

Yo, that's not an article that's the notebook. How awesome is that? This is for science is going.

Wow, that's very impressive! This kind of thing could really accelerate my research if I can get in the groove of it.

I just downloaded IJulia (IPython for Julia) and having been playing around with it.

This may be in the notebook or on Seaborn's homepage (http://stanford.edu/~mwaskom/software/seaborn/), but someone compared seaborn to pandas in that it provides a terse API to a great base package (matplotlib and numpy, respectively). I'm looking forward to using this more.

That's the first time I've ever heard the pandas API described as "terse".

Relative to creating your own ndarray, adding a time index, visualizing the data, etc?

Pandas is amazing and I use it all the time, but the API for things like hierarchical multi level indexing is very complex (it has to be - it's a very difficult thing).

The violin plots are cool and look like a great alternative to box plots, I had not seen those before. Sweet!

It is nice as it shows distributions, not only - a few statistics. However, some shapes of violin plots can be... distracting. ;)

I find it nicer than bare-bone matplotlib for plotting (one line of code per standard plot + very pleasant graphically).

I knew that it exists for some time (and I had it already installed) but only today I rediscovered it as a really useful tool for default plots in my data analysis / data exploration.

Very cool. I suspect many of these port to R as well, and it would be interesting to compare the code and approach.

scatter matrix is my goto high dimensional visualization.