matplotlib-users Mailing List for matplotlib

Hi Steve,
On Tuesday 21 February 2006 13:00, Steve Schmerler wrote:
> 1) Trying to export an .eps with using usetex = True:
>
> rcParams['text.usetex']=True; plot([1,2,3]); savefig("image.eps")
>
> I get an .eps where the axis numbers (are they called so?) are missing
> (looks like a bounding box problem). Only the axes and what's inside the
> plot box is there.
What mpl version are you using, and have you tried setting 
verbose=debug-annoying in your rc settings to see if any of the usetex 
external dependencies are complaining about anything?
> 2) Exporting .svg fails
There is no svg support for the usetex option.
Darren

Greetings,
I searched the mail archive for information on this question, and
didn't find anything. Hopefully this hasn't been addressed elsewhere.
I just spent a few hours tracking down a huge performance hit in
imshow(). I'm loading some image data via numarray's fromfile(),
stacking three imgaes to make an RGB, and showing it with imshow.
My image loading code looks like this:
def read_dib(filename):
 f =3D open(filename, "rb")
 fromfile(f, "UInt8", shape=3D(52, 1))
 im =3D fromfile(f, "UInt16", shape=3D(512, 512))
 if sys.byteorder =3D=3D "big":
=09im.byteswap()
 return im.astype('Float') / (2**12 - 1)
I load three images, and put them into a NxMx3 float numarray, and
pass that to imshow.
The performance drop is at this line in image.py:
 im =3D _image.fromarray(x, 0)
in AxesImage.make_image()
I eventually tracked it down to x's type being <class
'numarray.ma.MA.MaskedArray'>, which gets passed off to numarray's
NA_InputArray function, which chokes and calls setArrayFromSequence
(i.e., it's not using the fact that the input data is a numarray, and
is instead iterating over each element). This is taking around 30
seconds to convert a 512x512x3 image on a new Intel iMac.
I sped it up significantly by using this line instead:
 im =3D _image.fromarray(numerix.ma.filled(x,0), 0)
Not that that's a good idea (but it works for me for now).
There is no invalid data in my image. I'm not sure why it's ever
being turned into a masked array.
I'm using matplotlib 0.86.2 and numarray-1.5.1, on OS X 10.3.5 on an Intel =
iMac.
Any advice would be appreciated.
Thanks,
Ray Jones

The latest release of matplotlib was built against numpy 0.9.4.=20
Hopefully we will have a release soon to match up with the latest
numpy.
On 2/21/06, Gary <pa...@in...> wrote:
> John Hunter wrote:
>
> >
> > Humufr> This morning I install the cvs version of matplotlib and
> > Humufr> numpy and now I have a segfault when numpy is the
> > Humufr> numerical package define in matplotlibrc. It's seems to
> > Humufr> work fine with numarray.
> >
> >Make sure you have a clean install of both packages (rm -rf build
> >subdir in both and reinstall) and if you still get the segfault post a
> >minimum script run with --verbose-helpful ; this will give us the exact
> >version numbers of both the matplotlib and numpy packages.
> >
> >JDH
> >
> >
> >
>
> I just updated numpy to the latest version: 0.9.5, from the Windows
> binary installer. I have MPL 0.86.2, also binary installer.
>
> import pylab crashes python
> import matplotlib is ok
> backing up to numpy 0.9.4 works ok.
> does this issue belong to a numpy list?
>
>
> How exactly does one run a script with --verbose-helpful ? This does
> not seem to be a valid python switch.
>
> Here are the last lines prior to crash of
> python -v test.py
> where test.py contains one line: import pylab
>
> import matplotlib.axes # precompiled from
> c:\python24\lib\site-packages\matplotl
> ib\axes.pyc
> # c:\python24\lib\site-packages\matplotlib\artist.pyc matches
> c:\python24\lib\si
> te-packages\matplotlib\artist.py
> import matplotlib.artist # precompiled from
> c:\python24\lib\site-packages\matplo
> tlib\artist.pyc
> # c:\python24\lib\site-packages\matplotlib\transforms.pyc matches
> c:\python24\li
> b\site-packages\matplotlib\transforms.py
> import matplotlib.transforms # precompiled from
> c:\python24\lib\site-packages\ma
> tplotlib\transforms.pyc
> # c:\python24\lib\site-packages\matplotlib\_transforms.pyc matches
> c:\python24\l
> ib\site-packages\matplotlib\_transforms.py
> import matplotlib._transforms # precompiled from
> c:\python24\lib\site-packages\m
> atplotlib\_transforms.pyc
>
>
> -gary
>
>
> -------------------------------------------------------
> This SF.net email is sponsored by: Splunk Inc. Do you grep through log fi=
les
> for problems? Stop! Download the new AJAX search engine that makes
> searching your log files as easy as surfing the web. DOWNLOAD SPLUNK!
> http://sel.as-us.falkag.net/sel?cmd=3Dlnk&kid=3D103432&bid=3D230486&dat=
=3D121642
> _______________________________________________
> Matplotlib-users mailing list
> Mat...@li...
> https://lists.sourceforge.net/lists/listinfo/matplotlib-users
>

John Hunter wrote:
> This is an annoyance that is caused by the fact that the GTK install
> needs an X11 connection,
Is it really necessary to import GTK when installing MPL? why? I think 
getting rid of that requirement would be the best fix.
>Basically, you need to be able to
> make sure you can launch an X11 app (eg xterm) in the environment that
> you are running matplotlib install.
> 
> Perhaps some UNIX guru can give the proper incantation.
I've done it, but I'm not sure I remember how at the moment. But try 
something like:
xhost +root
or even:
xhost +
which I think lets anyone have access to the Xserver.
-Chris
-- 
Christopher Barker, Ph.D.
Oceanographer
 		
NOAA/OR&R/HAZMAT (206) 526-6959 voice
7600 Sand Point Way NE (206) 526-6329 fax
Seattle, WA 98115 (206) 526-6317 main reception
Chr...@no...

>>>>> "Jason" == Jason C Leach <jas...@gm...> writes:
 Jason> hi, I'm having trouble figuring out how to add a few more
 Jason> layers to a stacked bar graph. What I have so far is:
 Jason> p1 = axes.bar(ind, data['n'], width, color='r') p2
 Jason> = axes.bar(ind, data['l'], width, color='y',
 Jason> bottom=data['n']) p3 = axes.bar(ind, data['m'], width,
 Jason> color='b', bottom=data['l']) p4 = axes.bar(ind, data['h'],
 Jason> width, color='g', bottom=data['m'])
 Jason> But I don't think this is correct. I only get two layers
 Jason> (red and green). Does anyone know the proper method?
See examples/table_demo.py -- you need to keep a cumulative sum of the
bottom positions, eg
yoff = array([0.0] * len(colLabels)) # the bottom values for stacked bar chart
for row in xrange(rows):
 bar(ind, data[row], width, bottom=yoff, color=colours[row])
 yoff = yoff + data[row]
 cellText.append(['%1.1f' % (x/1000.0) for x in yoff])
JDH

hi,
I'm having trouble figuring out how to add a few more layers to a
stacked bar graph.
What I have so far is:
 p1 =3D axes.bar(ind, data['n'], width, color=3D'r')
 p2 =3D axes.bar(ind, data['l'], width, color=3D'y', bottom=3Ddata['=
n'])
 p3 =3D axes.bar(ind, data['m'], width, color=3D'b', bottom=3Ddata['=
l'])
 p4 =3D axes.bar(ind, data['h'], width, color=3D'g', bottom=3Ddata['=
m'])
But I don't think this is correct. I only get two layers (red and
green). Does anyone know the proper method?
Thanks,
Jason.
--
........................................
.... Jason C. Leach
.... PGP Key: 0x62DDDF75
.... Keyserver: gpg.mit.edu

John Hunter wrote:
>
> Humufr> This morning I install the cvs version of matplotlib and
> Humufr> numpy and now I have a segfault when numpy is the
> Humufr> numerical package define in matplotlibrc. It's seems to
> Humufr> work fine with numarray.
>
>Make sure you have a clean install of both packages (rm -rf build
>subdir in both and reinstall) and if you still get the segfault post a
>minimum script run with --verbose-helpful ; this will give us the exact
>version numbers of both the matplotlib and numpy packages.
>
>JDH
>
> 
>
I just updated numpy to the latest version: 0.9.5, from the Windows 
binary installer. I have MPL 0.86.2, also binary installer.
import pylab crashes python
import matplotlib is ok
backing up to numpy 0.9.4 works ok.
does this issue belong to a numpy list?
How exactly does one run a script with --verbose-helpful ? This does 
not seem to be a valid python switch.
Here are the last lines prior to crash of
python -v test.py
where test.py contains one line: import pylab
import matplotlib.axes # precompiled from 
c:\python24\lib\site-packages\matplotl
ib\axes.pyc
# c:\python24\lib\site-packages\matplotlib\artist.pyc matches 
c:\python24\lib\si
te-packages\matplotlib\artist.py
import matplotlib.artist # precompiled from 
c:\python24\lib\site-packages\matplo
tlib\artist.pyc
# c:\python24\lib\site-packages\matplotlib\transforms.pyc matches 
c:\python24\li
b\site-packages\matplotlib\transforms.py
import matplotlib.transforms # precompiled from 
c:\python24\lib\site-packages\ma
tplotlib\transforms.pyc
# c:\python24\lib\site-packages\matplotlib\_transforms.pyc matches 
c:\python24\l
ib\site-packages\matplotlib\_transforms.py
import matplotlib._transforms # precompiled from 
c:\python24\lib\site-packages\m
atplotlib\_transforms.pyc
-gary

1) Trying to export an .eps with using usetex = True:
rcParams['text.usetex']=True; plot([1,2,3]); savefig("image.eps")
I get an .eps where the axis numbers (are they called so?) are missing 
(looks like a bounding box problem). Only the axes and what's inside the 
plot box is there.
2) Exporting .svg fails:
In [9]: rcParams['text.usetex']=True; plot([1,2,3]); savefig("image.svg")
Out[9]: [<matplotlib.lines.Line2D instance at 0xb61d202c>]
---------------------------------------------------------------------------
exceptions.NotImplementedError Traceback (most 
recent call last)
/home/schmerler/<console>
/usr/lib/python2.3/site-packages/matplotlib/pylab.py in savefig(*args, 
**kwargs)
 813 def savefig(*args, **kwargs):
 814 fig = gcf()
--> 815 return fig.savefig(*args, **kwargs)
 816 if Figure.savefig.__doc__ is not None:
 817 savefig.__doc__ = _shift_string(Figure.savefig.__doc__)
/usr/lib/python2.3/site-packages/matplotlib/figure.py in savefig(self, 
*args, **kwargs)
 647 kwargs[key] = rcParams['savefig.%s'%key]
 648
--> 649 self.canvas.print_figure(*args, **kwargs)
 650
 651 def colorbar(self, mappable, cax=None,
/usr/lib/python2.3/site-packages/matplotlib/backends/backend_gtkagg.py 
in print_figure(self, filename, dpi, facecolor, edgecolor, orientation)
 109 agg = self.switch_backends(FigureCanvasAgg)
 110 try:
--> 111 agg.print_figure(filename, dpi, facecolor, 
edgecolor, orientation)
 112 except IOError, msg:
 113 error_msg_gtk('Failed to save\nError message: 
%s'%(msg,), self)
/usr/lib/python2.3/site-packages/matplotlib/backends/backend_agg.py in 
print_figure(self, filename, dpi, facecolor, edgecolor, orientation)
 473 from backend_svg import FigureCanvasSVG
 474 svg = self.switch_backends(FigureCanvasSVG)
--> 475 svg.print_figure(filename, dpi, facecolor, 
edgecolor, orientation)
 476 elif ext.find('ps')>=0 or ext.find('ep')>=0:
 477 from backend_ps import FigureCanvasPS # lazy import
/usr/lib/python2.3/site-packages/matplotlib/backends/backend_svg.py in 
print_figure(self, filename, dpi, facecolor, edgecolor, orientation)
 282 svgwriter = codecs.open( filename, 'w', 'utf-8' )
 283 renderer = RendererSVG(w, h, svgwriter)
--> 284 self.figure.draw(renderer)
 285 renderer.finish()
 286
/usr/lib/python2.3/site-packages/matplotlib/figure.py in draw(self, 
renderer)
 522
 523 # render the axes
--> 524 for a in self.axes: a.draw(renderer)
 525
 526 # render the figure text
/usr/lib/python2.3/site-packages/matplotlib/axes.py in draw(self, 
renderer, inframe)
 1436 if not self._axisbelow:
 1437 if self.axison and not inframe:
-> 1438 self.xaxis.draw(renderer)
 1439 self.yaxis.draw(renderer)
 1440
/usr/lib/python2.3/site-packages/matplotlib/axis.py in draw(self, 
renderer, *args, **kwargs)
 560 tick.set_label1(label)
 561 tick.set_label2(label)
--> 562 tick.draw(renderer)
 563 if tick.label1On:
 564 extent = tick.label1.get_window_extent(renderer)
/usr/lib/python2.3/site-packages/matplotlib/axis.py in draw(self, renderer)
 159 if self.tick2On: self.tick2line.draw(renderer)
 160
--> 161 if self.label1On: self.label1.draw(renderer)
 162 if self.label2On: self.label2.draw(renderer)
 163
/usr/lib/python2.3/site-packages/matplotlib/text.py in draw(self, renderer)
 856 if not self.get_visible(): return
 857 self.update_coords(renderer)
--> 858 self._mytext.draw(renderer)
 859 #bbox_artist(self._mytext, renderer, props={'pad':0}, 
fill=False)
 860 if self.get_dashlength() > 0.0:
/usr/lib/python2.3/site-packages/matplotlib/text.py in draw(self, renderer)
 346
 347 renderer.draw_tex(gc, x, y, line,
--> 348 self._fontproperties, angle)
 349 return
 350
/usr/lib/python2.3/site-packages/matplotlib/backend_bases.py in 
draw_tex(self, gc, x, y, s, prop, angle, ismath)
 320
 321 def draw_tex(self, gc, x, y, s, prop, angle, ismath='TeX!'):
--> 322 raise NotImplementedError
 323
 324 def draw_text(self, gc, x, y, s, prop, angle, ismath=False):
NotImplementedError:
cheers,
steve
-- 
Random number generation is the art of producing pure gibberish as 
quickly as possible.

For the benefit of the list, below is a working script that makes a
simple plot with a patch element, outputs the plot to a png, and
outputs an html file that contains an image map so that the patch
element is clickable.
Thanks for all the help,
Julius
#!/usr/bin/env python
from pylab import *
dots =3D 150
fig =3D figure(num=3D1,figsize=3D(6,4),dpi=3Ddots,facecolor=3D'w')
ax =3D fig.add_subplot(111)
pl =3D ax.plot(range(0,10),range(0,10),'b-')
box =3D ax.axvspan(3,6,fc=3D'm',alpha=3D0.5)
fig.savefig('test.png',dpi=3Ddots)
img_height =3D fig.get_figheight() * dots
b_verts =3D box.get_verts()
#image maps use left-x,left-y,right-x,right-y
verts =3D [(b_verts[1][0],b_verts[1][1]),(b_verts[3][0],b_verts[3][1])]
tverts =3D box.get_transform().seq_xy_tups(verts)
#image maps have y=3D0 on top
tverts =3D (tverts[0][0],img_height-tverts[0][1],tverts[1][0],img_height-tv=
erts[1][1])
file =3D open('test.html','w')
file.write('''<HTML>
<HEAD>
 <TITLE>Image Map Test</TITLE>
</HEAD>
<BODY>
<map name=3D"map">
 <area shape=3D"rect" coords=3D"%d,%d,%d,%d" href=3D"http://www.google.com=
">
</map>
<img border=3D2 src=3D"./test.png" usemap=3D"#map">
</BODY>
</HTML> ''' % tverts )
file.close()

>>>>> "Humufr" == Humufr <hu...@ya...> writes:
 Humufr> This morning I install the cvs version of matplotlib and
 Humufr> numpy and now I have a segfault when numpy is the
 Humufr> numerical package define in matplotlibrc. It's seems to
 Humufr> work fine with numarray.
Make sure you have a clean install of both packages (rm -rf build
subdir in both and reinstall) and if you still get the segfault post a
minimum script run with --verbose-helpful ; this will give us the exact
version numbers of both the matplotlib and numpy packages.
JDH

This morning I install the cvs version of matplotlib and numpy and now I 
have a segfault when numpy is the numerical package define in 
matplotlibrc. It's seems to work fine with numarray.

>>>>> "Julius" == Julius Lucks <jul...@gm...> writes:
 Julius> my question is what transform do I use to convert verts to
 Julius> pixel coordinates? ax.get_transform() does not work.
 Julius> Should I use pl.get_transform?
"Does not work" is not very descriptive :-(
Yes, you should use the polygon transform
You might want to look at the "draw" method of the
matplotlib.patches.Polygon class to see how it gets the data into
pixel coords
 verts = self.get_verts()
 tverts = self._transform.seq_xy_tups(verts)
Thus all you need to do outside the class is 
 trans = pl.get_transform()
 verts = pl.get_verts()
 tverts = trans.seq_xy_tups(verts)
 
Should work...
The documentation for the matplotlib.transforms module is decent, if
you want to take a look
 http://matplotlib.sourceforge.net/matplotlib.transforms.html
JDH

>>>>> "Stephen" == Stephen English <st...@se...> writes:
 Stephen> All, I am trying to render a set of data/date points with
 Stephen> plot_date. How can I get matplotlib to only plot the y
 Stephen> axis of this graph? I can turn both off with axis("off"),
 Stephen> but I can't seem to find more detailed instructions
 Stephen> anywhere. Also, although this is unimportant, it grates
You have to make the individual elements invisible. Something like
for label in ax.get_xticklabels():
 label.set_visible(False)
For more refined control, for each tick there are a host of visibility
properties you can set
 gridOn : a boolean which determines whether to draw the tickline
 tick1On : a boolean which determines whether to draw the 1st tickline
 tick2On : a boolean which determines whether to draw the 2nd tickline
 label1On : a boolean which determines whether to draw tick label
 label2On : a boolean which determines whether to draw tick label
for tick in ax.xaxis.get_major_ticks():
 tick.tick2On = False
and likewise for the minor ticks
 Stephen> somewhat to use matplotlib with the procedural (matlab
 Stephen> esque) instructions within my OO python code. Are there
 Stephen> any tutorials or other docs on plotting graphs through
 Stephen> the more pythonic libraries?
As the code above illustrates, matplotlib is an OO library. There is
no requirement to use it through the procedural pylab interface. Most
of my personal use, except for quick scripts and interactive sessions,
is OO.
See this FAQ
 http://matplotlib.sourceforge.net/faq.html#OO
and this tutorial
 http://matplotlib.sourceforge.net/leftwich_tut.txt
A nice hybrid approach is to use a minimum of pylab for figure
management only, and do everything else through OO
 import matplotlib.numerix as nx
 from pylab import figure, close, show # all you need
 
 x = nx.arange(0, 2.0, 0.01)
 y = nx.sin(2*nx.pi*x)
 fig1 = figure(1)
 fig2 = figure(2)
 
 ax1 = fig1.add_subplot(211)
 ax1.plot(x,y)
 ax1.set_xlabel('time (s)')
 ax1.set_ylabel('volts')
 ax1.set_title('a sinusoid')
 ax1.grid(True)
 
 # and more with fig2, etc....
 show()
 
Most of the action is in the Axes instance, and too a lesser extent
the Figure and Axis instances, so see the class documentation for the
figure, axes, and axis methods
 http://matplotlib.sourceforge.net/classdocs.html
Hope this helps,
JDH

>>>>> "Steve" == Steve Schmerler <el...@gm...> writes:
 Steve> ------------------------------------------------------------------------
 Steve> elcorto@ramrod:~/Install/Matplotlib/matplotlib-0.86.2$ sudo
 Steve> python setup.py install installing data to
 Steve> lib/python2.3/site-packages/matplotlib/mpl-data pygtk
 Steve> present but import failed Using default library and include
 Steve> directories for Tcl and Tk because a Tk window failed to
 Steve> open. You may need to define DISPLAY for Tk to work so
 Steve> that setup can determine where your libraries are located.
 Steve> Gtk-WARNING **: cannot open display:
 Steve> ------------------------------------------------------------------------
This is an annoyance that is caused by the fact that the GTK install
needs an X11 connection, and in some setups when you install as root
the root user does not have access to your display. I usually install
as suso: do you have sudo setup? Basically, you need to be able to
make sure you can launch an X11 app (eg xterm) in the environment that
you are running matplotlib install.
Perhaps some UNIX guru can give the proper incantation. In addition
to setting the DISPLAY as root, you may need to check your xhost
settings.
JDH

>>>>> "Tom" == Tom Denniston <tom...@al...> writes:
 Tom> I am generating a few hundred graphs and doing so takes on
 Tom> the order of about 10-15 minutes. Which seems to me rather
 Tom> slow. When I profile my code it identifies the calls to
 Tom> get_yticklabels and get_xticklabels as taking over 90% of the
 Tom> time. This seems strange but my calls to these functions are
 Tom> merely a sort round about way of setting the font size of the
 Tom> axis tick labels and suppressing the text for the
 Tom> xticklabels. Is there a more efficient and cleaner way to do
 Tom> this? artist.setp(axes.get_yticklabels(), visible=True,
 Tom> fontsize=7) artist.setp(axes.get_xticklabels(),
 Tom> visible=False)
This is a known performance bottleneck. There are two reasons it is
slow. Every tick label is handled as an independent object, when they
in most cases share most of their properties (font size, orientation)
and so could be better handled as a TextCollection, which does not
exist yet. The second reason is that the text layout engine is doing
layout for newline separated strings with an arbitrary rotation for
every tick label, which is almost never used. So some special case
optimizations to handle the no rotation, no newline text instances
(basically just bypass the layout machinery) would help a lot here.
Are you using matplotlib mathtext also, by chance? This slows things
down a bit too, though is better in recent versions.
JDH

I've had various versions of matplotlib from CVS installed on an 
Opteron (Red Hat LInux) here for a few weeks now. I had to modify 
setupext.py to look for various libraries in ../lib64 instead of /lib 
but matplotlib works very nicely most of the time, and I have started 
using it for real work.
One cosmetic problem: the buttons at the bottom of the plot window 
have their graphics messed up. They just look like red crosses. The 
floating xml works, so I know what the buttons do, and it doesn't 
inconvenience me --- but it looks bad.
Presumably some library is either out of date, incomplete, or not 
linked correctly. I am using the default TkAgg backend. Any thoughts 
on what I should look for would be appreciated.
BTW the matplotlib users_guide_0.83.2.pdf seems to be unreadable from 
my Mac. It appears blank in (Tiger) Safari, and preview cannot open 
the file: it gives 'File error'.
Regards, George Nurser.

Hi thanks for the help. I have a related problem. i had been using py2exe with matplotlib and it was working well. upgraded to 0.86 and now it doesn't work. I know this is because the installation directory is different now (no longer the share folder). So I thought I fixed it correctly but I get an error that indicates py2exe "can't find lib/site-packages/matplotlib/backends or it is not a regular file". I am assuming that it doesn't like 'backends' because it is a folder and not a file. I changed my setup code to this:
 
 data_files=[
 ("matplotlibdata",
 glob.glob(os.path.join(distutils.sysconfig.PREFIX, 'lib', 'site-packages','matplotlib', "*")))]
 
 setup(
 options = options,
 # The lib directory contains everything except the executables and the python dll.
 windows = [app],
 data_files = data_files,
 # use out build_installer class as extended py2exe build command
 cmdclass = {"py2exe": build_installer},
 )
 
 how might I change this to get it working properly? thanks!!
 
 Jeff
 
Charlie Moad <cw...@gm...> wrote: Recently the matplotlib data files have been moved into the matplotlib
module. You should be able to safely remove share/matplotlib.
On 2/19/06, Jeff Peery wrote:
> hello, I just upgraded from 0.84 to 0.86 on windows xp. I noticed that I
> have a couple different locations on my computere where it is installed.
> there is something in the python24/share and also
> python24/Lib/site-packages/
>
> what do I gotta do to make sure I am using the newest version. can I just
> delete all the matplot lib folders and reinstall the newest version?
>
> thakns!
>
>
> ________________________________
> Yahoo! Mail
> Use Photomail to share photos without annoying attachments.
>
>
		
---------------------------------
 
 What are the most popular cars? Find out at Yahoo! Autos