Hello all,
ggplot is a wrapper around matplotlib that mimics the way the R "ggplot" graphical module works.
It looks to duplicate "R" excellent "ggplot" graphic module into "matplotlib"
(see :
http://blog.yhathq.com/posts/ggplot-for-python.html)
If I understood well, it is based on :
* the gammar of graphic (
http://www.cs.uic.edu/~wilkinson/Publications/gpl.pdf)
* written by Leland Wilkinson
http://www.cs.uic.edu/~wilkinson/* and implemented in R by Hadley Wickham (
http://had.co.nz/)
==> I couldn't wait next winpython release to try it.
BUT, it requires the greatest and latest matplolib 1.3.1 (of october 10th, 2013)
==> as there is no new winpython official update since august 22nd, I made the effort to understand how to update myself.
It's dead simple indeed.
** Generic Procedure to update your winpython **
- for windows compiled modules, coming from "
http://www.lfd.uci.edu/~gohlke/pythonlibs/" :
* pick and download the module you want (choose the version corresponding to your installation : python 2.7 or 3.3, 32bit or 64 bit),
* launch "WinPython Control Panel.exe"
* click on "add packages" and select the executable you just downloaded
* click on the "install package(s)" button.
- for pure python modules :
* launch "WinPython Command Prompt.exe",
* type "pip install -U your_module"
** get ggplot procedure example (you may wait seeing ggplot version after 0.2.8 on pypi, if you're on python 3) ***
* download from "
http://www.lfd.uci.edu/~gohlke/pythonlibs" the matplotlib you need
(for example matplotlib-1.3.1.win32-py3.3.exe)
* launch "WinPython Control Panel.exe"
* click on "add packages" and select the executable you just downloaded
* click on the "install package(s)" button.
* launch "WinPython Command Prompt.exe",
* pip install ggplot
(or pip install -U ggplot, if you have an older version)
after that, fire up a winpython notebook and try this
**(start of code)************************
from ggplot import *
print (ggplot(aes(x='date', y='beef'), data=meat) + \
geom_point(alpha=0.3) + \
stat_smooth(colour="black", se=True))
plt.show(1)
**(end of code) ***********************
It looks it may become much simpler to do program graphics with matplotlib soon.