On 11/17/2010 01:28 PM, Ognjen Ilic wrote: > Thanks for the help. However, when I change the matplotlibrc file I > get the following message > "Bad key "path.simplify" on line 267 in > /HOME/.matplotlib/matplotlibrc. > You probably need to get an updated matplotlibrc file from > http://matplotlib.sf.net/matplotlibrc or from the matplotlib source > distribution" > > I'd do the update but I don't have root access on the server
It sounds like you have an old enough version of mpl that it does not support rc configuration of the path simplification. If so, your only option is to update mpl itself. It should be possible to build and install a local copy without root access, and put its location in your PYTHONPATH. Eric > > > On Wed, Nov 17, 2010 at 1:20 PM, Eric Firing<efir...@hawaii.edu> wrote: >> On 11/17/2010 07:35 AM, Ognjen Ilic wrote: >>> Hello all, >>> >>> I posted about this problem on another forum (with an image attachment) >>> http://python-forum.org/pythonforum/viewtopic.php?f=18&t=21951&p=99290#p99290 >>> >>> In the figure below white space that forms a trapezoid to the right >>> (slope then constant) and tear-shape white space to the left (due to >>> maxZ being too small) are expected and I have no issues with that. >>> The weird things are the sharp white shapes to the left. They are very >>> irregular and with a little different change in parameters they appear >>> at different places. >> >> This looks like the path simplification bug that was fixed some time >> before mpl 1.0. You can upgrade, or as a workaround you can turn off >> path simplification by putting >> >> path.simplify : False >> >> in your matplotlibrc file, or by using >> matplotlib.rcParams['path.simplify'] = False >> in your script. >> >> Eric >> >>> >>> Below are the relevant pieces of code >>> >>> Code: Select all >>> import matplotlib.pyplot as plt >>> import matplotlib.colors as colors >>> ... >>> z = numpy.transpose(z) >>> z = numpy.ma.masked_where(z<=0, z) >>> levels = numpy.linspace(0,maxZ,50) >>> >>> cset1=plt.contourf(x,y,z,levels,cmap=plt.get_cmap('jet',len(levels)-1)) >>> plt.colorbar(cset1) >>> fname = 'pic.png' >>> plt.savefig(fname) >>> >>> >>> >>> Above, z is the 2D array of values for different x and y (both lists >>> have 400 elements each). >>> Ths is really really confusing me. Any help is greatly appreciated! ------------------------------------------------------------------------------ Beautiful is writing same markup. Internet Explorer 9 supports standards for HTML5, CSS3, SVG 1.1, ECMAScript5, and DOM L2 & L3. Spend less time writing and rewriting code and more time creating great experiences on the web. Be a part of the beta today http://p.sf.net/sfu/msIE9-sfdev2dev _______________________________________________ Matplotlib-users mailing list Matplotlib-users@lists.sourceforge.net https://lists.sourceforge.net/lists/listinfo/matplotlib-users
