Healthy Algorithms

DEF CON 23 – Chris Rock – I Will Kill You

Here is a super-cool approach to making science more reproducible: the journal of visualized experiments http://www.jove.com/

Here is an example: http://www.jove.com/video/2065/combining-behavioral-endocrinology-experimental-economics

Here is a little recipe I cooked from today, a way to load a large Stata file into Python that you can cancel and still get something to play around with. I put it in an old StackOverflow question, so can somebody upvote it for me?

http://stackoverflow.com/a/31869752/1935494

Here is an answer to a question about zooming and tooltips in mpld3: http://stackoverflow.com/questions/31122064/tooltips-disappear-after-zoom-in-out-when-using-matplotlib-mpld3/31843371#31843371

I like the animation I made to answer it, but it is also something worth considering changing, because this interface *is* a bit confusing.

Today I learned about DataFrame.query and DataFrame.eval, which I first saw when I was looking over some students’ shoulders during code review last quarter. Very cool way to make things faster to run and to read: https://gist.github.com/aflaxman/f40ecd3c3eb33f5ab8d2

http://ioam.github.io/holoviews/Tutorials/Introduction with led me to https://ioam.github.io/param/ which could be useful for lots of scientific computation.

On my mind, due to some careful checking of results prior to submissions. Nothing substantive changed, but the numbers have to be right.

Correction
Data Errors in Results Section
JAMA. 2015;314(1):86 doi:10.1001/jama.2015.6076
Data Errors in Table 2
JAMA. 2015;314(1):86 doi:10.1001/jama.2015.6072
Data Errors in Table 3 and in Text
JAMA. 2015;314(1):86 doi:10.1001/jama.2015.6074
Incorrect Data in Table 4
JAMA. 2015;314(1):86 doi:10.1001/jama.2015.5464