Pandas cheat sheet

[changhiskhan]

common functions and recipes that fit onto a single sheet

[jreback]

looks that way

http://pandas.pydata.org/pandas-docs/dev/10min.html

@changhiskhan it's not 1 sheet but is there more 2 add?

[cpcloud]

@y-p should this be closed as well?

I think this is still relavent, 10min is great but the idea is to have something much
more dense, perferably a 1-2 page landscape pdf with some design behind it
and the ever-elusive pandas logo.

[cpcloud]

still think a panda punching honey badger is the way 2 go

If it must feature an animal famous for biting the head off other animals,
I'd opt for Ozzy Osbourne instead of a weird variation on Pepé Le Pew.

...But don't tell honey badger I said that.

[cpcloud]

even if i did he wouldn't give a hoot

[jess010]

I've started working on this. I've got my own ideas, but would appreciate some more feedback on what others feel this should look like before I get any deeper on this, as follows:

• Table style or dictionary style? Personally favor table, but both could be effective.
• Assumed knowledge? (ie assume some familiarity with pandas, numpy etc, or that they have not done the 10min exercise?) I would like to make this as simple as possible without making the more experienced ones want to claw out their eyeballs, but have a statement at the top of the cheatsheet that refers users to the 10min exercise if they are new to pandas.
• Include detail on how it works (e.g. a description of parameters), or what it does (e.g Series creates a 1D array), or both? My vote is for the what it does only - should be a quick refresher, not a detailed how-to.
• Contain links to more detailed docs? This will be useful, but would also be more cluttered.
• Where in the doc structure should this sit?

Random thought...Planet of the Pandas. Went to the movies and saw the preview for Dawn of the Apes. Incorporating Caesar's 'death unto you' expression and red war paint onto the face of a panda could make for an interesting logo.

[jreback]

would prob be in tutorials until complete

[jorisvandenbossche]

@jess010 Super you want to take this on!

See also #2552 for some initial work (maybe you can use some parts of it).

Were you thinking of something in the line of the R Reference Card? (http://cran.r-project.org/doc/contrib/Short-refcard.pdf) That is something I would love to see for pandas. You can even get the tex source of it if you are interested in that (http://cran.r-project.org/doc/contrib/Short-refcard.zip)

I think you can assume some knowledge, as this is always an overview to easily find/remember things again for people already using pandas (a bit), and not a real explanation for new users.

[qpleple]

Found this online, I don't know whose it is but it looks pretty good: https://docs.google.com/file/d/0ByIrJAE4KMTtS1ZmRldWSElRMmM/edit

[jreback]

@qpleple hmm, that does look good. Can you see if you can figure out who the author is an redirect them here?

[qpleple]

Here is the post on the author's blog
http://markthegraph.blogspot.fr/2014/01/pandas-dataframe-cheat-sheet-and-python.html