matplotlib-users Mailing List for matplotlib

Hi, I'm pretty unfamiliar with py2exe and I had been using it with matplotlib and it was working well. I upgraded to 0.86 and now it doesn't work. I know this is because the installation directory is different (no longer the share folder).  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

		
---------------------------------
Brings words and photos together (easily) with
 PhotoMail  - it's free and works with Yahoo! Mail.

Running : OS X 10.3.9  -  matplotlib 0.86 (Agg backend)   python 2.3  
Numeric 23.7

I have the following command sequence:

import Numeric
N = Numeric
from matplotlib import pylab
PL = pylab

x = N.arange(10.)
y = N.arange(10.)
y[-1] = 1.0e18

PL.plot(x,y,'r')
PL.xlim( 0.,7.)
PL.ylim( 0.,7.)
PL.savefig('plot1')
PL.clf()

PL.plot(x,y,'r--')
PL.xlim( 0.,7.)
PL.ylim( 0.,7.)
PL.savefig('plot2')
PL.clf()

Note that the y array has a screwy point  - 1.0e18.
What happens is that the first plot ('plot1') looks fine  but the code 
hangs on the savefig for the second plot - apparently nothing useful is 
happening -  it just hangs.
if I change the upper limits to 9 for plot 1, you can see a red filled 
polygon being drawn on the right hand side of the figure - do not know 
why ? With the limits of 7 all the nasty bits are not seen.
I suspect that the dashed option tries to produce a plot with a huge 
number of line segments and the code is overwhelmed - I am OK with 
overwhelmed but not with the code hanging. Something a bit more 
graceful should be done if possible.
Is there any thing I can check to avoid this situation?
In the actual case in which  this occurred I  tried to produce a plot  
after a lengthy computation. The computation had an outlier, It was a 
bit annoying to have the plot hang and thus I lost all the results to 
that point.

I could swear that this used to work(produce a plot in the second case) 
in an earlier version of matplotlib.  I had actually performed the 
computation and plot before and all went OK. I have the printed copy 
before me.

Thanks for any help.

--Jim

Eric Firing wrote:

> Jeff,
>
> I really blew it on that one--it looked so simple!  OK, I think this 
> version is correct; it will land in CVS in a few minutes.
>
>     def to_rgba(self, x, alpha=1.0):
>         '''Return a normalized rgba array corresponding to x.
>         If x is already an rgb or rgba array, return it unchanged.
>         '''
>         if hasattr(x, 'shape') and len(x.shape)>2: return x
>         x = ma.asarray(x)
>         x = self.norm(x)
>         x = self.cmap(x, alpha)
>         return x
> Eric
>

Eric:  Thanks for the quick fix - that seems to have done the job.

-Jeff

-- 
Jeffrey S. Whitaker         Phone  : (303)497-6313
Meteorologist               FAX    : (303)497-6449
NOAA/OAR/PSD  R/PSD1        Email  : Jef...@no...
325 Broadway                Office : Skaggs Research Cntr 1D-124
Boulder, CO, USA 80303-3328 Web    : http://tinyurl.com/5telg

Jeff,

I really blew it on that one--it looked so simple!  OK, I think this 
version is correct; it will land in CVS in a few minutes.

     def to_rgba(self, x, alpha=1.0):
         '''Return a normalized rgba array corresponding to x.
         If x is already an rgb or rgba array, return it unchanged.
         '''
         if hasattr(x, 'shape') and len(x.shape)>2: return x
         x = ma.asarray(x)
         x = self.norm(x)
         x = self.cmap(x, alpha)
         return x
Eric

Jeff Whitaker wrote:
> Eric Firing wrote:
> 
>> Ray,
>>
>> You tripped over a bug in cm.py: when I added masked array support, I 
>> put the "x = ma.asarray(x)" too early in the to_rgba() method of 
>> ScalarMappable.  It is fixed now in CVS.  Thanks for finding the problem.
>>
>> If you want to try the fix in your version, in place of your 
>> workaround, here is the revised method:
>>
>>    def to_rgba(self, x, alpha=1.0):
>>         # assume normalized rgb, rgba
>>         if len(x.shape)>2: return x
>>         x = ma.asarray(x)
>>         x = self.norm(x)
>>         x = self.cmap(x, alpha)
>>         return x
>>
>> Eric
>>
> 
> 
> Hi Eric:  This breaks some of the basemap demos which do this
> 
> CS = m.contour(x,y,hgt,15,linewidths=0.5,colors='k')
> 
> to draw solid black contour lines.  I now get
> 
> File "/sw/lib/python2.4/site-packages/matplotlib/cm.py", line 52, in 
> to_rgba
>    if len(x.shape)>2: return x
> AttributeError: 'list' object has no attribute 'shape'
> 
> It appears that colors now has to be an array?
> 
> -Jeff
> 

Eric Firing wrote:

> Ray,
>
> You tripped over a bug in cm.py: when I added masked array support, I 
> put the "x = ma.asarray(x)" too early in the to_rgba() method of 
> ScalarMappable.  It is fixed now in CVS.  Thanks for finding the problem.
>
> If you want to try the fix in your version, in place of your 
> workaround, here is the revised method:
>
>    def to_rgba(self, x, alpha=1.0):
>         # assume normalized rgb, rgba
>         if len(x.shape)>2: return x
>         x = ma.asarray(x)
>         x = self.norm(x)
>         x = self.cmap(x, alpha)
>         return x
>
> Eric
>


Hi Eric:  This breaks some of the basemap demos which do this

CS = m.contour(x,y,hgt,15,linewidths=0.5,colors='k')

to draw solid black contour lines.  I now get

 File "/sw/lib/python2.4/site-packages/matplotlib/cm.py", line 52, in 
to_rgba
    if len(x.shape)>2: return x
AttributeError: 'list' object has no attribute 'shape'

It appears that colors now has to be an array?

-Jeff

-- 
Jeffrey S. Whitaker         Phone  : (303)497-6313
Meteorologist               FAX    : (303)497-6449
NOAA/OAR/PSD  R/PSD1        Email  : Jef...@no...
325 Broadway                Office : Skaggs Research Cntr 1D-124
Boulder, CO, USA 80303-3328 Web    : http://tinyurl.com/5telg

Notable notes:
   - Built against numpy-0.9.5
   - A wealth of bugfixes

http://sourceforge.net/project/showfiles.php?group_id=3D80706

=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=
=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=
=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D
2006-02-22 Released 0.87

2006-02-21 Fixed portrait/landscape orientation in postscript backend - DSD

2006-02-21 fix bug introduced in yesterday's bug fix - SC

2006-02-20 backend_gtk.py FigureCanvasGTK.draw(): fix bug reported by
           David Tremouilles - SC

2006-02-20 Remove the "pygtk.require('2.4')" error from
           examples/embedding_in_gtk2.py - SC

2006-02-18 backend_gtk.py FigureCanvasGTK.draw(): simplify to use (rather t=
han
           duplicate) the expose_event() drawing code - SC

2006-02-12 Added stagger or waterfall plot capability to LineCollection;
           illustrated in examples/collections.py. - EF

2006-02-11 Massive cleanup of the usetex code in the postscript
backend. Possibly
           fixed the clipping issue users were reporting with older version=
s of
           ghostscript - DSD

2006-02-11 Added autolim kwarg to axes.add_collection.  Changed
           collection get_verts() methods accordingly. - EF

2006-02-09 added a temporary rc parameter text.dvipnghack, to allow
Mac users to get nice
           results with the usetex option. - DSD

2006-02-09 Fixed a bug related to setting font sizes with the usetex
option. - DSD

2006-02-09 Fixed a bug related to usetex's latex code. - DSD

2006-02-09 Modified behavior of font.size rc setting. You should
define font.size in pts,
           which will set the "medium" or default fontsize. Special
text sizes like axis
           labels or tick labels can be given relative font sizes like
small, large,
           x-large, etc. and will scale accordingly. - DSD

2006-02-08 Added py2exe specific datapath check again.  Also added new
           py2exe helper function get_py2exe_datafiles for use in py2exe
           setup.py scripts. - CM

2006-02-02 Added box function to pylab

2006-02-02 Fixed a problem in setupext.py, tk library formatted in unicode
           caused build problems - DSD

2006-02-01 Dropped TeX engine support in usetex to focus on LaTeX. - DSD

2006-01-29 Improved usetex option to respect the serif, sans-serif, monospa=
ce,
           and cursive rc settings. Removed the font.latex.package rc setti=
ng,
           it is no longer required - DSD

2006-01-29 Fixed tex's caching to include font.family rc information - DSD

2006-01-29 Fixed subpixel rendering bug in *Agg that was causing
           uneven gridlines - JDH

2006-01-28 Added fontcmd to backend_ps's RendererPS.draw_tex, to support ot=
her
           font families in eps output - DSD

2006-01-28 Added MaxNLocator to ticker.py, and changed contour.py to
           use it by default. - EF

2006-01-28 Added fontcmd to backend_ps's RendererPS.draw_tex, to support ot=
her
           font families in eps output - DSD

2006-01-27 Buffered reading of matplotlibrc parameters in order to allow
           'verbose' settings to be processed first (allows verbose.report
           during rc validation process) - DSD

2006-01-27 Removed setuptools support from setup.py and created a
           separate setupegg.py file to replace it. - CM

2006-01-26 Replaced the ugly datapath logic with a cleaner approach from
           http://wiki.python.org/moin/DistutilsInstallDataScattered.
           Overrides the install_data command. - CM

2006-01-24 Don't use character typecodes in cntr.c --- changed to use
           defined typenumbers instead. - TEO

2006-01-24 Fixed some bugs in usetex's and ps.usedistiller's dependency

2006-01-24 Added masked array support to scatter - EF

2006-01-24 Fixed some bugs in usetex's and ps.usedistiller's dependency
           checking - DSD

>>>>> "George" == George Nurser <ag...@no...> writes:

    George> I've had various versions of matplotlib from CVS installed
    George> on an Opteron (Red Hat LInux) here for a few weeks now. I
    George> had to modify setupext.py to look for various libraries in
    George> ../lib64 instead of /lib but matplotlib works very nicely
    George> most of the time, and I have started using it for real
    George> work.

    George> One cosmetic problem: the buttons at the bottom of the
    George> plot window have their graphics messed up. They just look
    George> like red crosses. The floating xml works, so I know what
    George> the buttons do, and it doesn't inconvenience me --- but it
    George> looks bad.

    George> Presumably some library is either out of date, incomplete,
    George> or not linked correctly. I am using the default TkAgg
    George> backend. Any thoughts on what I should look for would be
    George> appreciated.

We used to see something like this on OSX with the WX backend.  What
version of Tk are you using?  Could this be a byte order or byte size
problem?

The code that loads the buttons is in
matplotlib.backends.backend_tkagg

    def _Button(self, text, file, command):
        file = os.path.join(rcParams['datapath'], file)
        im = Tk.PhotoImage(master=self, file=file)
        b = Tk.Button(
            master=self, text=text, padx=2, pady=2, image=im, command=command)
        b._ntimage = im
        b.pack(side=Tk.LEFT)
        return b


you may want to read up on the tk PhotoImage and Button classes and
see if there is some option that makes this work.  I don't have access
to the opteron platform to test on.  The tk backend uses the ppm files
for the icons (eg home.ppm).  Do any of the other icon image formats
that ship with matplotlib (xpm, png, svg) work for you?  You can test
this by editing the _init_toolbar code in backend_tkagg.py

        self.bHome = self._Button( text="Home", file="home.ppm",
                                   command=self.home)

and trying different extensions.

Thanks,
JDH

On Wednesday 22 February 2006 7:32 am, Bruce wrote:
> Hi group,
>
> I wanted to try to usetex for my matplotlib plots, and thought that I
> had everything neccessary installed, but it doesn`t work. Perhaps
> someone in the group have some idea what is wrong here? Im using
> MikTex on xp and have the dvipng-miktex pakage installed. thx

Have you read the wiki? http://www.scipy.org/Wiki/Cookbook/Matplotlib/UsingTex

An intermediate dvi file is missing, probably because the latex executable is 
not on your path.

Darren

Hi group,

I wanted to try to usetex for my matplotlib plots, and thought that I
had everything neccessary installed, but it doesn`t work. Perhaps
someone in the group have some idea what is wrong here? Im using
MikTex on xp and have the dvipng-miktex pakage installed. thx

  File "C:\\Python23\\lib\\site-packages\\matplotlib\\text.py", line
334, in draw
    bbox, info =3D self._get_layout(renderer)
  File "C:\\Python23\\lib\\site-packages\\matplotlib\\text.py", line
179, in _get_layout
    w,h =3D renderer.get_text_width_height(
  File "C:\\Python23\\lib\\site-packages\\matplotlib\\backends\\backend_agg=
.py",
line 237, in get_text_width_height
    Z =3D self.texmanager.get_rgba(s, size, dpi, rgb)
  File "C:\\Python23\\lib\\site-packages\\matplotlib\\texmanager.py",
line 276, in get_rgba
    pngfile =3D self.make_png(tex, dpi, force=3DFalse)
  File "C:\\Python23\\lib\\site-packages\\matplotlib\\texmanager.py",
line 113, in make_png
    dvifile =3D self.make_dvi(tex)
  File "C:\\Python23\\lib\\site-packages\\matplotlib\\texmanager.py",
line 105, in make_dvi
    shutil.move(dvitmp, dvifile)
  File "C:\\Python23\\lib\\shutil.py", line 174, in move
    copy2(src,dst)
  File "C:\\Python23\\lib\\shutil.py", line 83, in copy2
    copyfile(src, dst)
  File "C:\\Python23\\lib\\shutil.py", line 38, in copyfile
    fsrc =3D open(src, 'rb')
IOError: [Errno 2] No such file or directory:
'a8cf192be8520ffec23612a6a3a0ee29.dvi
>>>

Christopher Barker wrote:
> 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
> 


I've replied to only to John, not the list. Here is the mail again:

================================================================================================

John Hunter wrote:

 >>>>>> "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.


Exactly this was the problem. I was logged in from machine A on machine 
B as user (ssh -X B) and wanted to install on B. So what I did on B was:

(as user):
user@B:<mpl_path>$ python setup.py build  # this works


(via sudo):
user@B:<mpl_path>$ sudo python setup.py install
python setup.py install installing data to
lib/python2.3/site-packages/matplotlib/mpl-data
pygtk present but import failed
Using default library and include directories for Tcl and Tk because a 
Tk window failed to open.  You may need to define DISPLAY for Tk to work 
so that setup can determine where your libraries are located.

Gtk-WARNING **: cannot open display:


(becoming su):
user@B:<mpl_path>$ su
user@B:<mpl_path># python setup.py install
X11 connection rejected because of wrong authentication.
The application 'setup.py' lost its connection to the display 
localhost:10.0;
most likely the X server was shut down or you killed/destroyed
the application.


 > 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.


Right, I can't open an X app on B via sudo or beeing su. Unfortunalelly 
I don't understand enough of the X-permission/xhost stuff to get this 
right :)

 > Perhaps some UNIX guru can give the proper incantation.


He did. Finally I logged in as root (ssh -X root@B) and then everthing
went fine.

================================================================================================


Additionally, if you want to install directly on A or B (not beeing 
logged in via ssh) just open a root terminal and install as usual. sudo 
and su won't do it.

cheers,
steve

-- 
Random number generation is the art of producing pure gibberish as 
quickly as possible.

Darren Dale wrote:
> 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?
> 

I'm using 0.86.2 (from sourceforge).

This script

--------------------------------------
from pylab import *
from matplotlib import verbose, use

use('GTKAgg')
verbose.level = 'debug-annoying'

rcParams['text.usetex']=True
plot([1,2,3])
savefig("image.eps")
--------------------------------------

produces

---------------------------------------------------------------------------
schmerler@bach:~$ python test.py
FigureCanvasAgg.print_figure
FigureCanvasAgg.draw
RendererAgg.__init__
RendererAgg._get_agg_font
         findfont failed Lucida Grande
         findfont failed Verdana
         findfont failed Geneva
         findfont failed Lucida
         findfont found Bitstream Vera Sans, normal, normal 500, normal, 
12.0
findfont returning 
/usr/lib/python2.3/site-packages/matplotlib/mpl-data/Vera.ttf
RendererAgg.points_to_pixels
RendererAgg.points_to_pixels
RendererAgg._get_agg_font
RendererAgg.points_to_pixels
RendererAgg.points_to_pixels
RendererAgg._get_agg_font
RendererAgg.points_to_pixels
RendererAgg.points_to_pixels
RendererAgg._get_agg_font
RendererAgg._get_agg_font
RendererAgg._get_agg_font
RendererAgg.points_to_pixels
RendererAgg.points_to_pixels
RendererAgg._get_agg_font
RendererAgg.points_to_pixels
RendererAgg.points_to_pixels
RendererAgg._get_agg_font
RendererAgg.points_to_pixels
RendererAgg.points_to_pixels
RendererAgg._get_agg_font
RendererAgg._get_agg_font
matplotlib data path /usr/lib/python2.3/site-packages/matplotlib/mpl-data
latex -interaction=nonstopmode "/tmp/15f8552d172b5a41c557b4ad371e3782.tex"
This is e-TeX, Version 3.14159-2.1 (Web2C 7.4.5)
entering extended mode
(/tmp/15f8552d172b5a41c557b4ad371e3782.tex
LaTeX2e <2001/06/01>
Babel <v3.7h> and hyphenation patterns for american, french, german, 
ngerman, b
ahasa, basque, catalan, croatian, czech, danish, dutch, finnish, greek, 
iceland
ic, irish, italian, latin, magyar, norsk, norsk, portuges, romanian, 
russian, s
lovak, slovene, spanish, swedish, turkish, ukrainian, nohyphenation, loaded.
(/usr/share/texmf/tex/latex/koma-script/scrartcl.cls
Document Class: scrartcl 2003/01/31 v2.9n LaTeX2e KOMA document class
(/usr/share/texmf/tex/latex/koma-script/scrlfile.sty
Package scrlfile, 2003/01/31 v2.9n LaTeX2e KOMA package
                   Copyright (C) Markus Kohm

) (/usr/share/texmf/tex/latex/base/size11.clo)
(/usr/share/texmf/tex/latex/koma-script/typearea.sty
Package typearea, 2003/01/31 v2.9n LaTeX2e KOMA package
                   Copyright (C) Frank Neukam, 1992-1994
                   Copyright (C) Markus Kohm, 1994-2002

)) (/usr/share/texmf/tex/latex/misc/type1cm.sty)
(/usr/share/texmf/tex/latex/misc/psfrag.sty
(/usr/share/texmf/tex/latex/graphics/graphics.sty
(/usr/share/texmf/tex/latex/graphics/trig.sty)
(/usr/share/texmf/tex/latex/config/graphics.cfg)
(/usr/share/texmf/tex/latex/graphics/dvips.def)))
(/usr/share/texmf/tex/latex/graphics/graphicx.sty
(/usr/share/texmf/tex/latex/graphics/keyval.sty))
(/usr/share/texmf/tex/latex/graphics/color.sty
(/usr/share/texmf/tex/latex/config/color.cfg)
(/usr/share/texmf/tex/latex/graphics/dvipsnam.def))
No file 15f8552d172b5a41c557b4ad371e3782.aux.
<15f8552d172b5a41c557b4ad371e3782.eps> [1]
(./15f8552d172b5a41c557b4ad371e3782.aux) )
Output written on 15f8552d172b5a41c557b4ad371e3782.dvi (1 page, 2352 bytes).
Transcript written on 15f8552d172b5a41c557b4ad371e3782.log.


dvips -R -T 12.750000in,18.030000in -o 
"/tmp/15f8552d172b5a41c557b4ad371e3782.ps" 
"/tmp/15f8552d172b5a41c557b4ad371e3782.dvi"

This is dvips(k) 5.92b Copyright 2002 Radical Eye Software 
(www.radicaleye.com)
' TeX output 2006.02.22:1254' -> /tmp/15f8552d172b5a41c557b4ad371e3782.ps
<texc.pro><psfrag.pro><f7b6d320.enc><texps.pro><special.pro><color.pro>.
<cmr10.pfb>[1<15f8552d172b5a41c557b4ad371e3782.eps>]

gs -dBATCH -dNOPAUSE -dSAFER -r6000 
-sDEVICE=epswrite -dLanguageLevel=2 -dEPSFitPage 
  -sOutputFile="/tmp/15f8552d172b5a41c557b4ad371e3782.eps" 
"/tmp/15f8552d172b5a41c557b4ad371e3782.ps"
GPL Ghostscript 8.01 (2004-01-30)
Copyright (C) 2004 artofcode LLC, Benicia, CA.  All rights reserved.
This software comes with NO WARRANTY: see the file PUBLIC for details.
---------------------------------------------------------------------------

It complains about not finding some Agg fonts, could this be it??

Trying to export an .ps rather than an .eps produces a .ps where 
everything is visible but the image is a whole A4 page.

>>2) Exporting .svg fails
> 
> 
> There is no svg support for the usetex option.

Oh too bad, would have been nice.

cheers,
steve

-- 
Random number generation is the art of producing pure gibberish as 
quickly as possible.

Ray,

You tripped over a bug in cm.py: when I added masked array support, I 
put the "x = ma.asarray(x)" too early in the to_rgba() method of 
ScalarMappable.  It is fixed now in CVS.  Thanks for finding the problem.

If you want to try the fix in your version, in place of your workaround, 
here is the revised method:

    def to_rgba(self, x, alpha=1.0):
         # assume normalized rgb, rgba
         if len(x.shape)>2: return x
         x = ma.asarray(x)
         x = self.norm(x)
         x = self.cmap(x, alpha)
         return x

Eric

Thouis (Ray) Jones wrote:
> 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 = open(filename, "rb")
>     fromfile(f, "UInt8", shape=(52, 1))
>     im = fromfile(f, "UInt16", shape=(512, 512))
>     if sys.byteorder == "big":
> 	im.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 = _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 = _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
> 
> 
> -------------------------------------------------------
> This SF.net email is sponsored by: Splunk Inc. Do you grep through log files
> 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=k&kid3432&bid#0486&dat1642
> _______________________________________________
> Matplotlib-users mailing list
> Mat...@li...
> https://lists.sourceforge.net/lists/listinfo/matplotlib-users