matplotlib-users Mailing List for matplotlib

>>>>> "Robert" == Robert Kern <rob...@gm...> writes:
    Robert> Well, that's not the right answer, either.

    Robert> In [18]: numpy.array([1])/0.0 Out[18]: array([ inf])

    Robert> But I think the *real* right answer is this:

    Robert> In [15]: numpy.inf Out[15]: inf

OK, now I am thoroughly confused.  Take a look at the original
traceback

    from _nc_imports import nx, inf, infinity, Infinity
  File "C:\Python24\Lib\site-packages\matplotlib\numerix\_nc_imports.py",
line 25, in ?
    inf = infty = infinity = Infinity = (array([1])/0.0)[0]
TypeError: unsupported operand type(s) for /: 'numpy.ndarray' and 'float'

He is importing from _nc_imports, which is triggered by Numeric, not
numpy.  numpy lives in _sp_imports.  _nc_imports does the following

  from Numeric import array
  #...snip, snip
  inf = infty = infinity = Infinity = (array([1])/0.0)[0]

How in the world is he getting the error message pointing to an
numpy.ndarray in _nc_imports?

JDH

John Hunter wrote:
>>>>>>"Sturla" == Sturla Molden <stu...@me...> writes:
> 
>     Sturla> "C:\Python24\Lib\site-packages\matplotlib\numerix\_nc_imports.py",
>     Sturla> line 25, in ?  inf = infty = infinity = Infinity =
>     Sturla> (array([1])/0.0)[0] TypeError: unsupported operand type(s)
>     Sturla> for /: 'numpy.ndarray' and 'float'
>     >>>>
> 
> Are you sure about your numpy version?  I am not seeing this on linux
> 
> In [1]: import numpy
> 
> In [4]: numpy.__version__
> Out[4]: '0.9.5'
> 
> In [3]: numpy.ndarray([1])/0.0
> Out[3]: array([             -inf])
> 
> Could you run these commands on your system?

Well, that's not the right answer, either.

In [18]: numpy.array([1])/0.0
Out[18]: array([              inf])

But I think the *real* right answer is this:

In [15]: numpy.inf
Out[15]: inf

-- 
Robert Kern
rob...@gm...

"I have come to believe that the whole world is an enigma, a harmless enigma
 that is made terrible by our own mad attempt to interpret it as though it had
 an underlying truth."
  -- Umberto Eco

>>>>> "Sturla" == Sturla Molden <stu...@me...> writes:
    Sturla> "C:\Python24\Lib\site-packages\matplotlib\numerix\_nc_imports.py",
    Sturla> line 25, in ?  inf = infty = infinity = Infinity =
    Sturla> (array([1])/0.0)[0] TypeError: unsupported operand type(s)
    Sturla> for /: 'numpy.ndarray' and 'float'
    >>>>

Are you sure about your numpy version?  I am not seeing this on linux

In [1]: import numpy

In [4]: numpy.__version__
Out[4]: '0.9.5'

In [3]: numpy.ndarray([1])/0.0
Out[3]: array([             -inf])

Could you run these commands on your system?

JDH

>>>>> "James" == James Boyle <bo...@ll...> writes:

    James> Does the capability exist in matplotlib to use patterns
    James> rather than colors to fill polygons, contours etc.?  This
    James> ability comes in handy when one wants to distinguish
    James> various regions in a plot but needs to restrict the plot to
    James> black and white.  The page charges are sometimes much
    James> larger for colored graphics and reproduction costs are
    James> likewise reduced for B/W.

Recent versions of matplotlib support hatching in the PS backend only
-- it would be nice to support general pattern fills in *Agg and
others too....

import matplotlib
matplotlib.use('PS')
from pylab import figure

fig = figure()
ax = fig.add_subplot(111)
bars = ax.bar(range(1,5), range(1,5), color='gray', ecolor='black')

patterns = ('/', '+', 'x', '\\')
for bar, pattern in zip(bars, patterns):
     bar.set_hatch(pattern)
fig.savefig('hatch4.ps')

Hi All,
I am new to matplotlib. I really appreciate help with building it!

When I downloaded the source: matplotlib-0.87.1.tar.gz
and tried build in the cygwin environment (CYGWIN_NT-5.1 YZhao
1.5.19(0.150/4/2) 2006-01-20 13:28 i686 Cygwin)

I got the following error:
$ python setup.py build
     6 [main] python2.4 41284 C:\GNU\bin\python2.4.exe: *** fatal
error - unable to remap C:\GNU\usr\X11R6\bin\cygXcursor-1.dll to same
address as parent(0x19130000) !=3D 0x19160000
  3052 [main] python 41844 child_copy: loaded dll data write copy
failed, 0x63104000..0x63105150, done 0, windows pid 2287012, Win32
error 5
Traceback (most recent call last):
 File "setup.py", line 203, in ?
   build_gdk(ext_modules, packages, NUMERIX)
 File "/cygdrive/e/tmp/matplotlib-0.87.1/setupext.py", line 838, in build_g=
dk
   add_pygtk_flags(module)
 File "/cygdrive/e/tmp/matplotlib-0.87.1/setupext.py", line 191, in
add_pygtk_flags
   pygtkIncludes =3D getoutput('pkg-config --cflags-only-I pygtk-2.0').spli=
t()
 File "/cygdrive/e/tmp/matplotlib-0.87.1/setupext.py", line 113, in getoutp=
ut
   ret =3D  os.popen(s).read().strip()
OSError: [Errno 11] Resource temporarily unavailable

Thanks a lot in advance!

Does the capability exist in matplotlib to use patterns rather than 
colors to fill polygons, contours etc.?
This ability comes in handy when one wants to distinguish various 
regions in a plot but needs to restrict the plot to black and white.
The page charges are sometimes much larger for colored graphics and 
reproduction costs are likewise reduced for B/W.

--Jim

On Wednesday 08 March 2006 07:22, Randewijk P-J <pjr...@su...> wrote:
> if sys.platform == 'win32': gs_exe = 'gswin32c'
> else: gs_exe = 'gs'
> command = '%s -dBATCH -dNOPAUSE -sDEVICE=bbox "%s"'
> % (gs_exe, tmpfile)

Thank you for the report, this has been fixed.

Darren

On Wednesday 08 March 2006 16:48, you wrote:
> >>>>> "Darren" == Darren Dale <dd...@co...> writes:
>
>     Darren> Any objections to making the backend rc setting
>     Darren> case-insensitive?
>
> I think it is a good idea -- just put a .lower() in the relevant
> places.

Unfortunately, it wasnt as easy as that. I had some trouble with plots and 
toolbars not rendering (although the window would render), and the 
debug-annoying output wasnt helpful, so I gave up on this route. 

Instead, I made a validate_backend function that validates "gtkagg" regardless 
of the case, and returns the properly formatted "GTKAgg". I tested with 
backend_driver, I tried setting the backend with the use function, and with 
the -d flag at runtime, all tests passed. Changes in svn.

Darren

Hi All,
I am new to matplotlib. I really appreciate help with building it!

When I downloaded the source: matplotlib-0.87.1.tar.gz
and tried build in the cygwin environment (CYGWIN_NT-5.1 YZhao
1.5.19(0.150/4/2) 2006-01-20 13:28 i686 Cygwin)

I got the following error:
$ python setup.py build
      6 [main] python2.4 41284 C:\GNU\bin\python2.4.exe: *** fatal
error - unable to remap C:\GNU\usr\X11R6\bin\cygXcursor-1.dll to same
address as parent(0x19130000) !=3D 0x19160000
   3052 [main] python 41844 child_copy: loaded dll data write copy
failed, 0x63104000..0x63105150, done 0, windows pid 2287012, Win32
error 5
Traceback (most recent call last):
  File "setup.py", line 203, in ?
    build_gdk(ext_modules, packages, NUMERIX)
  File "/cygdrive/e/tmp/matplotlib-0.87.1/setupext.py", line 838, in build_=
gdk
    add_pygtk_flags(module)
  File "/cygdrive/e/tmp/matplotlib-0.87.1/setupext.py", line 191, in
add_pygtk_flags
    pygtkIncludes =3D getoutput('pkg-config --cflags-only-I pygtk-2.0').spl=
it()
  File "/cygdrive/e/tmp/matplotlib-0.87.1/setupext.py", line 113, in getout=
put
    ret =3D  os.popen(s).read().strip()
OSError: [Errno 11] Resource temporarily unavailable

Thanks a lot in advance!

>>>>> "Imara" == Imara Jarrett <im...@gm...> writes:

    Imara> Greetings, I am trying to make a scatter plot with each
    Imara> marker having an individual color.

    Imara> What am I doing wrong?

    Imara> x = [ 4, 5, 6] y = [ 7, 8, 9] size = [ 6,7,8] color = [
    Imara> 'r', 'b', 'g'] scatter( x, y, s = size, c = color) show()

    Imara> I get the following error:

    Imara> AttributeError: 'list' object has no attribute 'astype'

    Imara> Any suggestions would be appreciated!

This is fixed in the svn repository.  For now, make sure that your
color argument is an array, not a list.

JDH

Greetings,

I am trying to make a scatter plot with each marker having an individual
color.

What am I doing wrong?

x =3D [ 4, 5, 6]
y =3D [ 7, 8, 9]
size =3D [ 6,7,8]
color =3D [ 'r', 'b', 'g']
scatter( x, y, s =3D size, c =3D color)
show()

I get the following error:

AttributeError: 'list' object has no attribute 'astype'

Any suggestions would be appreciated!

John Hunter <jdh...@ac...> writes:

> What you need to do is find out which cm* font file contains the
> Omega you want, and what the glyph index of Omega is.

The simple way to do this is to write a TeX file like

    \documentclass{article}
    \pagestyle{empty}
    \begin{document}
    $\Gamma \Delta \Theta \Lambda \Xi \Pi \Sigma \Upsilon \Phi \Psi \Omega$
    \end{document}

run it through LaTeX and then do dvitype -output-level=2 foo.dvi,
which reveals the font and the codes:

    114: fntdef1 7: cmr10---loaded at size 655360 DVI units
    135: fntnum7
    136: setchar0
    137: setchar1
    138: setchar2
    139: setchar3
    140: setchar4
    141: setchar5
    142: setchar6
    143: setchar7
    144: setchar8
    145: setchar9
    146: setchar10

So the upper-case non-slanted Greek letters occupy the first 11
positions of cmr10. For the letters not listed, I guess you are
supposed to use the similar-looking Latin letters.

-- 
Jouni

On Wed, 08 Mar 2006, John Hunter <jdh...@ac...> wrote...

> 
> It looks like matplotlib finally made it into debian
> 
>   http://packages.debian.org/unstable/python/python2.3-matplotlib
> 
> Hey, now we are officially unstable.  Something to celebrate!

That's awesome news! Congrats to JDH and all the others who've put in so 
much hard work for this great tool.

Cheers,

Tim

> 
> There has been a lot of interest in getting mpl into debian for some
> time, but because the install and prereqs are so damn complicated
> (look at the list!), it took quite a while before someone put together
> a package that would satisfy the debian mavens.
> 
> JDH
> 
> 
> -------------------------------------------------------
> This SF.Net email is sponsored by xPML, a groundbreaking scripting language
> that extends applications into web and mobile media. Attend the live webcast
> and join the prime developer group breaking into this new coding territory!
> http://sel.as-us.falkag.net/sel?cmd=lnk&kid=110944&bid=241720&dat=121642
> _______________________________________________
> Matplotlib-users mailing list
> Mat...@li...
> https://lists.sourceforge.net/lists/listinfo/matplotlib-users
> 
`-

It looks like matplotlib finally made it into debian

  http://packages.debian.org/unstable/python/python2.3-matplotlib

Hey, now we are officially unstable.  Something to celebrate!

There has been a lot of interest in getting mpl into debian for some
time, but because the install and prereqs are so damn complicated
(look at the list!), it took quite a while before someone put together
a package that would satisfy the debian mavens.

JDH

I have problems getting matplotlib to work on Windows (ActivePython
2.4.2). After installing the precompiled binary packages for NumPy 0.9.5
and Matplotlib 0.87.1, I get this error message:

>>> import pylab as p
Traceback (most recent call last):
  File "<pyshell#4>", line 1, in ?
    import pylab as p
  File "C:\Python24\Lib\site-packages\pylab.py", line 1, in ?
    from matplotlib.pylab import *
  File "C:\Python24\Lib\site-packages\matplotlib\pylab.py", line 196, in ?
    import cm
  File "C:\Python24\Lib\site-packages\matplotlib\cm.py", line 5, in ?
    import colors
  File "C:\Python24\Lib\site-packages\matplotlib\colors.py", line 33, in ?
    from numerix import array, arange, take, put, Float, Int, where, \
  File "C:\Python24\Lib\site-packages\matplotlib\numerix\__init__.py",
line 61, in ?
    from _nc_imports import nx, inf, infinity, Infinity
  File "C:\Python24\Lib\site-packages\matplotlib\numerix\_nc_imports.py",
line 25, in ?
    inf = infty = infinity = Infinity = (array([1])/0.0)[0]
TypeError: unsupported operand type(s) for /: 'numpy.ndarray' and 'float'
>>>


Is there an easy fix for this?


Sturla Molden

John,
I think this contains the answer to your question: 
http://www.aptech.com/support_faq.html#vb
I have not tried it. 
Wish I could be of more help, 
Alan 
PS Had trouble with your email address ...

"numpy" is the third generation of Numeric/numarray, it is a different 
package.

http://numeric.scipy.org/

You can go get it and install it, or change your matplotlibrc to select 
Numeric or numarray

-Chris


Chris Seberino wrote:
> import pylab
> 
> causes it to try to import numpy
> 
> but 'numpy' is not found in numarray or numeric packages!?!
> 
> Any ideas?
> 
> 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...