You can find the project here:
Goggle form sign in link:
Note the new venue. HiVelocity (https://www.hivelocity.net/) has graciously offered to host our meetings! (Thanks Harvey). You may see the address referred to as suite 700 but the 500 is not a typo.
I've written a data table "auto-analyzer" and I've been working to generalize it for any data set. If you have any non-proprietary data, bring a file and let's see what happens. It's set-up for xlsx files at this point but conversion is easy.
For this Meetup, I'll start with a brief description of the output and code and then we'll try to run it on new files. You win if your file throws a fatal exception and I have to open it up to see what went wrong. Then, if there's time, we can dig into the code a bit if anyone wants to see how it works (hint: it relies heavily on PANDAS and MATPLOTLIB).
The only known constraint is that it be a "flat file." By that I mean a simple 2-D data table. It doesn't even really need column headers but the results are more interesting if there are. Oh yeah, don't get too carried away with size either. Not sure what "too large" is. I know it can safely handle 100 or so columns and upwards of 200K rows. Much more than that and run time might make for a boring presentation if nothing else.
I'm really curious what people will come up with. Plus, this will hopefully build a fire under me to clean up the code a bit and get it on GitHub.
Bring your laptop! Or at least some data on a thumb drive...