matplotlib-users Mailing List for matplotlib (Page 6)

There are a couple things about legend that I'm finding a little
irksome.  Is there some better way to do this?

1) if you have a contour, legend() wants to add all the contours to
the list.  calling contour(...,label='_nolegend_') doesn't seem to
help.

I'm trying to plot a bunch of different types of markers on top of a
contour plot, and put just the markers in the legend.  The workaround
seems to be
    contour(. . . )
    p1 = plot( . . .)
    p2 = plot(. . .)
    . . .
    pN = plot(. . .)

    legend((p1,p2,...,pN), ('Legend1', 'Legend2', ... , 'LegendN'))
    Basically I have to save the plot data and repeat myself later.

2) The '_nolegend_' trick doesn't work for things in the simple call
style for legend()

     legend('_nolegend_', 'Data 1', 'Data 2')

That makes a legend containing the string "_nolegend_".

My first hunch to do that before looking at the docs at all was to put
in a None for a legend I wanted to leave off.  That's raises an error
though.  Might be nice if it was made a legal way to omit something
from the legend.

--bb

Hi Jeff,
I want to display some pollutants concentrations saved in shapefiles on map=
s=20
displayed in 3D. You can have a look of what I need here:
http://rockware.com/catalog/pages/arcgis3dfeat1.html, in the last image.

Le Jeudi 12 Avril 2007 16:40, Jeff Whitaker a =E9crit=A0:
> Lionel Roubeyrie wrote:
> > Hi all,
> > not sure it's possible, I can't find any example but I ask : is there a
> > way to make some 3D maps with Basemap, like we can do with ArcScene
> > (ESRI), mixing rasters or shapefiles with TINs? If no, do you know a way
> > to do so with python?
> > Thanks
>
> Lionel:  I don't know, but I honestly doubt it.  Matplotlib is
> fundamentally a 2-D package, there is a bit of mostly unsupported 3-D
> stuff that has been bolted on.   Can you be more specific about what you
> mean though?  Perhaps a link to an example?
>
> I have had good luck with ncvtk (http://ncvtk.sourceforge.net/), but I
> think it only really works if your data is global and in a netCDF file.
>
> -Jeff

=2D-=20
Lionel Roubeyrie - lro...@li...
Chag=E9 d'=E9tudes et de maintenance
LIMAIR - la Surveillance de l'Air en Limousin
http://www.limair.asso.fr

belinda thom wrote:
> 
[...]
> 
> Here's more info (redundant?) regarding what I've tried, in case it 
> clarifies. (Ultimately, I will obviously need to upgrade to 0.9, at 
> which point I can report back on if this fixed the problem on my 
> machine, but I can't do that until after the semester is over, mid May). 
> I'm running MacPython 2.4.4. I've got matplotlib 0.87.7 w/TkAgg. I'm 
> using IPython 0.7.4.svn.r2010, called with the -pylab flag. I'm running 
> the tests reported here on a G5 (I've also seen them on a G4), using OS 
> X 10.4.8.

Belinda,

I was not paying close enough attention and did not realize that you are 
all the way back at 0.87.7.  I think that trying to do any kind of 
piecemeal updating of that to get 3d working decently is an exercise in 
futility; a lot of things have changed since then, some problems in 3d 
have been fixed but it still can't be considered operational, and Tim's 
recent work has been post-0.90, so upgrading to that is still not likely 
to be very satisfactory.  A new release is planned soon--maybe next 
week?--and that would certainly get you closer.  But overall, I think 
that trying to use mpl 3d functionality for a class this term is not a 
good idea.  There is a lot you can do in 2d with 0.87.7; I think you 
would really be best advised to find a way to make that adequate for the 
purposes of your class this term.

The easiest way to see some of the things that have changed is to go to
http://matplotlib.svn.sourceforge.net/viewvc/matplotlib/trunk/matplotlib/ 
and browse the archive, especially lib/matplotlib, where the pure python 
parts are.  0.87.7 was released at revision 2835.  You can also get an 
overview from CHANGELOG and API_CHANGES.

Eric

On Apr 12, 2007, at 11:02 AM, Eric Firing wrote:

> belinda thom wrote:
>
>> Me neither :-). I will try and get permission to upgrade axis.py w/ 
>> my quick fix commenting out the kwargs in line 2131 of axes.py, as  
>> this is the quick option and I need something ASAP. You seem to  
>> know quite a bit about matplotlib. If you have any idea what this  
>> hack might break, I'd appreciate your insight so I know what to be  
>> wary of.
>
> Your "fix" is rendering the scalex and scaley kwargs of Axes.plot  
> ineffective, so that both axes will be autoscaled regardless of the  
> value of these kwargs.  It might have no practical consequence for  
> what you are doing.  The larger concern is that I can't figure out  
> how making that change would be needed--the error message you show  
> doesn't make sense to me, given what I see in the code, which means  
> I don't understand something critical--and in fact this "fix"  
> doesn't seem to be needed on my system.  So, maybe this particular  
> problem really has been fixed since 0.9, and Tim is referring to  
> some other sort of problem in the 3d code. (I am simply assuming he  
> is working from svn, but I don't know this for a fact.)


> All very confusing and unsatisfactory.

I agree.

Here's more info (redundant?) regarding what I've tried, in case it  
clarifies. (Ultimately, I will obviously need to upgrade to 0.9, at  
which point I can report back on if this fixed the problem on my  
machine, but I can't do that until after the semester is over, mid  
May). I'm running MacPython 2.4.4. I've got matplotlib 0.87.7 w/ 
TkAgg. I'm using IPython 0.7.4.svn.r2010, called with the -pylab  
flag. I'm running the tests reported here on a G5 (I've also seen  
them on a G4), using OS X 10.4.8.

In [2]: import matplotlib.axes3d as P3
In [3]: f = P.figure()
In [4]: ax = P3.Axes3D(f)
In [5]: ax.plot3d([1,2,3],[1,2,3],[1,2,3],'ro-')
In [6]: P.show()
------------------------------------------------------------------------ 
---
exceptions.TypeError                                 Traceback (most  
recent call last)

/Users/bthom/belinda/mills/aicourse/material/week12/lec/examples/ 
<ipython console>

/Library/Frameworks/Python.framework/Versions/2.4/lib/python2.4/site- 
packages/matplotlib/axes3d.py in plot3D(self, xs, ys, zs, *args,  
**kwargs)
     488     def plot3D(self, xs, ys, zs, *args, **kwargs):
     489         had_data = self.has_data()
--> 490         lines = Axes.plot(self, xs,ys, *args, **kwargs)
     491         if len(lines)==1:
     492             line = lines[0]

/Library/Frameworks/Python.framework/Versions/2.4/lib/python2.4/site- 
packages/matplotlib/axes.py in plot(self, *args, **kwargs)
    2129         lines = [line for line in lines] # consume the  
generator
    2130
-> 2131         self.autoscale_view(scalex=scalex, scaley=scaley)
    2132         #self.autoscale_view()#scalex=scalex, scaley=scaley)
    2133         return lines

TypeError: autoscale_view() got an unexpected keyword argument 'scalex'

In [6]:

NOTE COMMENTED OUT LINE 2132. NOW I'LL DO THE SAME THING (FIRST  
RESTARTING IPYTHON) W/THE OTHER LINE COMMENTED OUT INSTEAD,

No error occurs and the figure plots as one would expect.

If there is anything else I can do to clear things up, please let me  
know.

I've got a list of other things that don't work the way I'd expect  
that I will post soon. Perhaps you can give me a link to where to  
find the changes made since 0.87.7 first, as I'm new to reading that  
kind of thing. Hopefully when Tim tries this stuff out on his end,  
we'll know more.

Thanks,

--b

Stephen Uhlhorn wrote:
> Just to be clear, the installation order is:
> 
> 1) install python2.5 from macpython.
> 2) remove wxagg.so
> 3) install wxpython frim macpython
> 4) install numpy/mpl from macpython

that's out of order. wxagg.so is part of mpl, so:

1) install python2.5 from macpython.
2) install wxpython frim macpython
3) install numpy/mpl from macpython
4) remove wxagg.so

-Chris


-- 
Christopher Barker, Ph.D.
Oceanographer

Emergency Response Division
NOAA/NOS/OR&R            (206) 526-6959   voice
7600 Sand Point Way NE   (206) 526-6329   fax
Seattle, WA  98115       (206) 526-6317   main reception

Chr...@no...

Just to be clear, the installation order is:

1) install python2.5 from macpython.
2) remove wxagg.so
3) install wxpython frim macpython
4) install numpy/mpl from macpython

Correct?
-stephen

On 4/12/07, Christopher Barker <Chr...@no...> wrote:
> Stephen Uhlhorn wrote:
> > Do I need to manually remove _wxagg.so in order to use the new packages?
>
> if you want to use them with wxPython 2.8.*, yes.
>
> They *should* work with 2.6.*, with _wxagg.so, but I haven't tested
> that. This should all be better with the next release.
>
> -Chris
>
>
> --
> Christopher Barker, Ph.D.
> Oceanographer
>
> Emergency Response Division
> NOAA/NOS/OR&R            (206) 526-6959   voice
> 7600 Sand Point Way NE   (206) 526-6329   fax
> Seattle, WA  98115       (206) 526-6317   main reception
>
> Chr...@no...
>

belinda thom wrote:

> Me neither :-). I will try and get permission to upgrade axis.py w/my 
> quick fix commenting out the kwargs in line 2131 of axes.py, as this is 
> the quick option and I need something ASAP. You seem to know quite a bit 
> about matplotlib. If you have any idea what this hack might break, I'd 
> appreciate your insight so I know what to be wary of.

Your "fix" is rendering the scalex and scaley kwargs of Axes.plot 
ineffective, so that both axes will be autoscaled regardless of the 
value of these kwargs.  It might have no practical consequence for what 
you are doing.  The larger concern is that I can't figure out how making 
that change would be needed--the error message you show doesn't make 
sense to me, given what I see in the code, which means I don't 
understand something critical--and in fact this "fix" doesn't seem to be 
needed on my system.  So, maybe this particular problem really has been 
fixed since 0.9, and Tim is referring to some other sort of problem in 
the 3d code. (I am simply assuming he is working from svn, but I don't 
know this for a fact.)  All very confusing and unsatisfactory.

Eric

Grant,

Grant Edwards wrote:
> OK, I'm on my third day of fighting with matplotlib.  I'm
> finally able to get py2exe to produce a "dist" directory, but
> it _still_ doesn't work:
>
>   Traceback (most recent call last):
>     File "surfacedit.py", line 3, in ?
>     File "wxmpl.pyc", line 23, in ?
>     File "matplotlib\__init__.pyc", line 715, in ?
>     File "matplotlib\__init__.pyc", line 273, in wrapper
>     File "matplotlib\__init__.pyc", line 360, in _get_data_path
>   RuntimeError: Could not find the matplotlib data files
>
> Here's the setup file I'm using:
>
>
> from distutils.core import setup
> import py2exe
> import matplotlib
>
> setup(windows=[{"script":"surfacedit.py"}],
>       data_files=[('',['../win32/gnuplot/bin/wgnuplot.exe',
>                        '../win32/gnuplot/bin/pgnuplot.exe',
>                        '../win32/gnuplot/bin/wgnuplot_pipes.exe']
>                   ) + matplotlib.get_py2exe_datafiles()],
>   
the above line does not look correct.  I.e. the '+' in there, what is it
for?

I have a folder structure of
dist/lib
dist/prog

so this one works for me
  ("lib\\matplotlibdata", mpfiles), # use this line if use zipfile option

if one does not use the py2exe zipfile option, which I use, then this
should work
  matplotlib.get_py2exe_datafiles(), # or this one if you don't use the
zipfile option

If one uses the zipfile option then all of the matplotlib stuff is under:
dist/lib
and it expects to have the data stuff in
dist/lib

otherwise it wants it in
dist

>       options={'py2exe': {'packages': ['matplotlib.numerix','pytz'],
>                           'dll_excludes': ['tcl84.dll', 'tk84.dll', 'wxmsw26uh_vc.dll'],
>                           'excludes': ["Tkconstants", "Tkinter", "tcl", '_gtkagg', '_tkagg']}}
>       )
>
> Is there any current, correct, documentation on how to bundle a
> matplotlib app?      
>
>   

I don't know, but I think the setup.py I posted to this list for
"embedding_in_wx4.py" is pretty accurate and at least for me it does
build a matplotlib app.

Werner

Hi Werner,

I am using matplotlib version 0.90.0 for python 2.4.

Thank you,

Archana.

On 4/12/07, Grant Edwards <gr...@vi...> wrote:
>
> On 2007-04-01, Werner F. Bruhin <wer...@fr...> wrote:
> > Hi Archana,
> >
> > Sometimes py2exe can't figure out what needs to be included.  In these
> > cases one creates entries in the packages section to force the inclusion
> > of one or multiple packages.
> >
> > Archana Ganesan wrote:
> >> Hi all,
> >> ...
> >>
> >> options = {"py2exe": {"compressed": 1,
> >>                      "optimize": 2,
> >>                      "packages": ["encodings",
> >> ##                                   "kinterbasdb",
> >>                                   "pytz.zoneinfo.UTC",
> >>                                   #" matplotlib.numerix",
> >>
> >
> > You need to activate/un-comment the matplotlib.numerix line.  Also note
> > that if you only include UTC and you use the timezone stuff in matplot
> > then a user NOT using UTC, i.e. another timezone will have a problem, it
> > is therefore better to just include all of pytz.
>
> Why does matplotlib need timezone data??
>
> --
> Grant Edwards                   grante             Yow!  Yow! Those people
>                                   at               look exactly like
> Donnie
>                                visi.com            and Marie Osmond!!
>
>
> -------------------------------------------------------------------------
> Take Surveys. Earn Cash. Influence the Future of IT
> Join SourceForge.net's Techsay panel and you'll get the chance to share
> your
> opinions on IT & business topics through brief surveys-and earn cash
> http://www.techsay.com/default.php?page=join.php&p=sourceforge&CID=DEVDEV
> _______________________________________________
> Matplotlib-users mailing list
> Mat...@li...
> https://lists.sourceforge.net/lists/listinfo/matplotlib-users
>

OK, I'm on my third day of fighting with matplotlib.  I'm
finally able to get py2exe to produce a "dist" directory, but
it _still_ doesn't work:

  Traceback (most recent call last):
    File "surfacedit.py", line 3, in ?
    File "wxmpl.pyc", line 23, in ?
    File "matplotlib\__init__.pyc", line 715, in ?
    File "matplotlib\__init__.pyc", line 273, in wrapper
    File "matplotlib\__init__.pyc", line 360, in _get_data_path
  RuntimeError: Could not find the matplotlib data files

Here's the setup file I'm using:


from distutils.core import setup
import py2exe
import matplotlib

setup(windows=[{"script":"surfacedit.py"}],
      data_files=[('',['../win32/gnuplot/bin/wgnuplot.exe',
                       '../win32/gnuplot/bin/pgnuplot.exe',
                       '../win32/gnuplot/bin/wgnuplot_pipes.exe']
                  ) + matplotlib.get_py2exe_datafiles()],
      options={'py2exe': {'packages': ['matplotlib.numerix','pytz'],
                          'dll_excludes': ['tcl84.dll', 'tk84.dll', 'wxmsw26uh_vc.dll'],
                          'excludes': ["Tkconstants", "Tkinter", "tcl", '_gtkagg', '_tkagg']}}
      )

Is there any current, correct, documentation on how to bundle a
matplotlib app?      

-- 
Grant Edwards                   grante             Yow!  My EARS are GONE!!
                                  at               
                               visi.com            

On 2007-04-01, Werner F. Bruhin <wer...@fr...> wrote:
> Hi Archana,
>
> Sometimes py2exe can't figure out what needs to be included.  In these 
> cases one creates entries in the packages section to force the inclusion 
> of one or multiple packages.
>
> Archana Ganesan wrote:
>> Hi all,
>> ...
>>
>> options = {"py2exe": {"compressed": 1,
>>                      "optimize": 2,
>>                      "packages": ["encodings",
>> ##                                   "kinterbasdb",
>>                                   "pytz.zoneinfo.UTC",
>>                                   #" matplotlib.numerix",
>>   
>
> You need to activate/un-comment the matplotlib.numerix line.  Also note 
> that if you only include UTC and you use the timezone stuff in matplot 
> then a user NOT using UTC, i.e. another timezone will have a problem, it 
> is therefore better to just include all of pytz.

Why does matplotlib need timezone data??

-- 
Grant Edwards                   grante             Yow!  Yow! Those people
                                  at               look exactly like Donnie
                               visi.com            and Marie Osmond!!

Stephen Uhlhorn wrote:
> Do I need to manually remove _wxagg.so in order to use the new packages?

if you want to use them with wxPython 2.8.*, yes.

They *should* work with 2.6.*, with _wxagg.so, but I haven't tested 
that. This should all be better with the next release.

-Chris


-- 
Christopher Barker, Ph.D.
Oceanographer

Emergency Response Division
NOAA/NOS/OR&R            (206) 526-6959   voice
7600 Sand Point Way NE   (206) 526-6329   fax
Seattle, WA  98115       (206) 526-6317   main reception

Chr...@no...

I've found a way to configure my compiler. After setting CC=3Dgcc I was
able to complete the compilation. But when I test it :

python anim.py

The import of the numpy version of the _transforms module,
_ns_transforms, failed.  This is is either because numpy was unavailable
when matplotlib was compiled, because a dependency of _ns_transforms
could not be satisfied, or because the build flag for this module was
turned off in setup.py.  If it appears that _ns_transforms was not
built, make sure you have a working copy of numpy and then re-install
matplotlib. Otherwise, the following traceback gives more details:

Traceback (most recent call last):
  File "anim.py", line 19, in <module>
    import pylab as p
  File "/Produits/tmp/rla/AIX/lib/python2.5/site-packages/pylab.py",
line 1, in <module>
    from matplotlib.pylab import *
  File
"/Produits/tmp/rla/AIX/lib/python2.5/site-packages/matplotlib/pylab.py",
line 203, in <module>
    from axes import Axes, PolarAxes
  File
"/Produits/tmp/rla/AIX/lib/python2.5/site-packages/matplotlib/axes.py",
line 14, in <module>
    import artist
  File
"/Produits/tmp/rla/AIX/lib/python2.5/site-packages/matplotlib/artist.py"
, line 4, in <module>
    from transforms import identity_transform
  File
"/Produits/tmp/rla/AIX/lib/python2.5/site-packages/matplotlib/transforms
.py", line 223, in <module>
    from _transforms import Value, Point, Interval, Bbox, Affine
  File
"/Produits/tmp/rla/AIX/lib/python2.5/site-packages/matplotlib/_transform
s.py", line 17, in <module>
    from matplotlib._ns_transforms import *
ImportError: No module named _ns_transforms

Of course, I have numpy installed and tested.

During the compilation, I get this error (using gcc-4.1.1):

/usr/local/lib/python2.5/config/ld_so_aix cc_r
-bI:/usr/local/lib/python2.5/config/python.exp
build/temp.aix-5.3-2.5/src/_na_transforms.o
build/temp.aix-5.3-2.5/src/mplutils.o
build/temp.aix-5.3-2.5/CXX/IndirectPythonInterface.o
build/temp.aix-5.3-2.5/CXX/cxx_extensions.o
build/temp.aix-5.3-2.5/CXX/cxxsupport.o
build/temp.aix-5.3-2.5/CXX/cxxextensions.o -L/usr/local/lib
-L/opt/freeware/lib -L/Produits/publics/powerpc.AIX.5.3/lib
-L/Produits/publics/powerpc.AIX.5.3/gcc/3.1/lib
-L/Produits/publics/powerpc.AIX.5.2/zlib/1.1.4/lib
-L/Produits/tmp/JL/AIX-5.2/TestInstall/lib -lstdc++ -lm -o
build/lib.aix-5.3-2.5/matplotlib/_na_transforms.so
ld: 0711-317 ERROR: Undefined symbol: _ZTISt9exception
ld: 0711-317 ERROR: Undefined symbol: ._Unwind_Resume
ld: 0711-317 ERROR: Undefined symbol:
._ZSt29_Rb_tree_insert_and_rebalancebPSt18_Rb_tree_node_baseS0_RS_
ld: 0711-317 ERROR: Undefined symbol:
._ZSt18_Rb_tree_decrementPSt18_Rb_tree_node_base
ld: 0711-317 ERROR: Undefined symbol:
._ZSt18_Rb_tree_incrementPSt18_Rb_tree_node_base
ld: 0711-317 ERROR: Undefined symbol: _ZNSs4_Rep20_S_empty_rep_storageE
ld: 0711-317 ERROR: Undefined symbol:
._ZN9__gnu_cxx18__exchange_and_addEPVii
ld: 0711-317 ERROR: Undefined symbol: _ZTISt11logic_error
ld: 0711-317 ERROR: Undefined symbol: .__cxa_guard_acquire
ld: 0711-317 ERROR: Undefined symbol: .__cxa_guard_release
ld: 0711-317 ERROR: Undefined symbol: .__cxa_guard_abort
ld: 0711-345 Use the -bloadmap or -bnoquiet option to obtain more
information. building 'matplotlib._ns_transforms' extension

Lionel Roubeyrie wrote:
> Hi all,
> not sure it's possible, I can't find any example but I ask : is there a way to 
> make some 3D maps with Basemap, like we can do with ArcScene (ESRI), mixing 
> rasters or shapefiles with TINs? If no, do you know a way to do so with 
> python?
> Thanks
>
>   
Lionel:  I don't know, but I honestly doubt it.  Matplotlib is 
fundamentally a 2-D package, there is a bit of mostly unsupported 3-D 
stuff that has been bolted on.   Can you be more specific about what you 
mean though?  Perhaps a link to an example?

I have had good luck with ncvtk (http://ncvtk.sourceforge.net/), but I 
think it only really works if your data is global and in a netCDF file.

-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

Do I need to manually remove _wxagg.so in order to use the new packages?

Thanks Chris!
-stephen

On 4/11/07, Christopher Barker <Chr...@no...> wrote:
> Hi all,
>
> I've got the MPL 0.90.0 installer on pythonmac working OK with:
>
> Python2.5
> wxPython2.8.3
>
> I accomplished this by removing:
>
> /Library/Frameworks/Python.framework/Versions/2.5/lib/python2.5/site-packages/matplotlib/backends/_wxagg.so
>
> Which disables the accelerator that doesn't work with wxPython 2.8
>
> What we really need to do is get Ken's changes into a release, but I
> have my immediate needs met.
>
> -Chris
>
>
>
>
> --
> Christopher Barker, Ph.D.
> Oceanographer
>
> Emergency Response Division
> NOAA/NOS/OR&R            (206) 526-6959   voice
> 7600 Sand Point Way NE   (206) 526-6329   fax
> Seattle, WA  98115       (206) 526-6317   main reception
>
> Chr...@no...
>
> -------------------------------------------------------------------------
> Take Surveys. Earn Cash. Influence the Future of IT
> Join SourceForge.net's Techsay panel and you'll get the chance to share your
> opinions on IT & business topics through brief surveys-and earn cash
> http://www.techsay.com/default.php?page=join.php&p=sourceforge&CID=DEVDEV
> _______________________________________________
> Matplotlib-users mailing list
> Mat...@li...
> https://lists.sourceforge.net/lists/listinfo/matplotlib-users
>

On Tue, 2007-04-10 at 19:03 +0300, Jouni K. Sepp=C3=A4nen wrote:
> David Fokkema <dfo...@il...> writes:
>=20
> > I can't think of an application where you have bins of different
> > widths and you want to center the values...
>=20
> Actually, now that I think about it, there is not enough information
> in the bin centers to know the widths of the bins if they may vary.
> For example, if your bin edges are (2, 4, 8, 16, 32) or (1, 5, 7, 17,
> 31), you get the same bin centers (3, 6, 12, 24). Perhaps it's best to
> disallow variable-width bins when align=3D'center'.

Of course! Nice example, ;-) I've changed the documentation strings.
What do you think of this patch? Shall I send it upstream as a bug
report with attached patch?

David

--- matplotlib/axes.py.orig	2007-04-12 09:52:47.000000000 +0200
+++ matplotlib/axes.py	2007-04-12 14:26:08.000000000 +0200
@@ -4137,19 +4137,21 @@
         n/(len(x)*dbin)
=20
         align =3D 'edge' | 'center'.  Interprets bins either as edge
-        or center values
+	or center values. If 'center', the bins are interpreted as equally
+	sized.
=20
         orientation =3D 'horizontal' | 'vertical'.  If horizontal, barh
         will be used and the "bottom" kwarg will be the left edges.
=20
         width: the width of the bars.  If None, automatically compute
-        the width.
+	the width. If align =3D 'center', the bins are interpreted as
+	equally sized.
=20
         kwargs are used to update the properties of the
         hist bars
         """
         if not self._hold: self.cla()
-        n, bins =3D matplotlib.mlab.hist(x, bins, normed)
+        n, bins =3D matplotlib.mlab.hist(x, bins, normed, align)
         if width is None: width =3D 0.9*(bins[1]-bins[0])
         if orientation =3D=3D 'horizontal':
             patches =3D self.barh(bins, n, height=3Dwidth, left=3Dbottom,
align=3Dalign)
--- matplotlib/mlab.py.orig	2007-04-12 09:52:47.000000000 +0200
+++ matplotlib/mlab.py	2007-04-12 14:25:30.000000000 +0200
@@ -597,7 +597,7 @@
    #S =3D -1.0*asum(p*log(p))
    return S
=20
-def hist(y, bins=3D10, normed=3D0):
+def hist(y, bins=3D10, normed=3D0, align=3D'edge'):
     """
     Return the histogram of y with bins equally sized bins.  If bins
     is an array, use the bins.  Return value is
@@ -605,7 +605,12 @@
=20
     If normed is False, return the counts in the first element of the
     return tuple.  If normed is True, return the probability density
-    n/(len(y)*dbin)
+    n/(len(y)*dbin). If normed is True, the bins are interpreted as
+    equally sized.
+
+    align =3D 'edge' | 'center'.  Interprets bins either as edge
+    or center values. If 'center', the bins are interpreted as equally
+    sized.
=20
     If y has rank>1, it will be raveled
     Credits: the Numeric 22 documentation
@@ -626,11 +631,16 @@
         dy =3D (ymax-ymin)/bins=20
         bins =3D ymin + dy*arange(bins)
=20
+    if align =3D=3D 'center':
+	hw =3D .5*(bins[1]-bins[0])
+	nbins =3D [x-hw for x in bins]
+    else:
+	nbins =3D bins
=20
-    n =3D searchsorted(sort(y), bins)
+    n =3D searchsorted(sort(y), nbins)
     n =3D diff(concatenate([n, [len(y)]]))
     if normed:
-       db =3D bins[1]-bins[0]
+       db =3D nbins[1]-nbins[0]
        return 1/(len(y)*db)*n, bins
     else:
        return n, bins

Eric,

On Apr 12, 2007, at 12:16 AM, Eric Firing wrote:

> belinda thom wrote:
>> Thanks Eric!
>> I'm kind of afraid to upgrade to the SVN version b/c I fear  
>> something else might break (I've yet to install matplotlib from  
>> source b/c of a few nagging Mac OS X issues).
>> What would you recommend?
>
> I don't know; I see from your earlier message that Tim Leslie says  
> some of the things you are trying to do are still broken, and his  
> patches are what John committed, so svn is certainly no better than  
> what Tim is working with now.

Ick. I hadn't noticed the version he was using. Should I file a bug  
report? and if so, where?

> I don't think that anything has changed since 0.90 in such a way as  
> to require changing both the compiled parts and the pure python  
> parts, so you could try simply replacing t

This was the info that I was missing in order to understand how to  
proceed. Not sure how I'd replace "just the python part" ---  
presumably something manual? But I'm glad to hear this is a possible  
option.

> There's also the issue of having to upgrade student machines for  
> which I'm not admin. Grrr.

 > That certainly makes the prospect of doing any kind of upgrade  
less attractive.  I don't have a good solution.

Me neither :-). I will try and get permission to upgrade axis.py w/my  
quick fix commenting out the kwargs in line 2131 of axes.py, as this  
is the quick option and I need something ASAP. You seem to know quite  
a bit about matplotlib. If you have any idea what this hack might  
break, I'd appreciate your insight so I know what to be wary of.

Thx,
--b

Hi all,
not sure it's possible, I can't find any example but I ask : is there a way=
 to=20
make some 3D maps with Basemap, like we can do with ArcScene (ESRI), mixing=
=20
rasters or shapefiles with TINs? If no, do you know a way to do so with=20
python?
Thanks

=2D-=20
Lionel Roubeyrie - lro...@li...
Chag=C3=A9 d'=C3=A9tudes et de maintenance
LIMAIR - la Surveillance de l'Air en Limousin
http://www.limair.asso.fr

I'm trying to compile matplotlib under AIX 5.3 (I've installed
ActivePython), here is what I get:

/usr/local/python2.5/config/ld_so_aix cc_r
-bI:/usr/local/python2.5/config/python.exp
build/temp.aix-5.3-2.5/src/agg.o
build/temp.aix-5.3-2.5/agg23/src/agg_trans_affine.o
build/temp.aix-5.3-2.5/agg23/src/agg_path_storage.o
build/temp.aix-5.3-2.5/agg23/src/agg_bezier_arc.o
build/temp.aix-5.3-2.5/agg23/src/agg_vcgen_dash.o
build/temp.aix-5.3-2.5/agg23/src/agg_vcgen_stroke.o
build/temp.aix-5.3-2.5/agg23/src/agg_rasterizer_scanline_aa.o
build/temp.aix-5.3-2.5/agg23/src/agg_curves.o -lstdc++ -lm -o
build/lib.aix-5.3-2.5/matplotlib/_agg.so
ld: 0706-006 Cannot find or open library file: -l stdc++
        ld:open(): No such file or directory
building 'matplotlib._na_transforms' extension
creating build/temp.aix-5.3-2.5/CXX
cc_r -DNDEBUG -O -Isrc -I. -I/opt/freeware/include
-I/Produits/publics/powerpc.AIX.5.3/include
-I/Produits/publics/powerpc.AIX.5.3/gcc/3.1/include -I.
-I/usr/local/include/python2.5 -c src/_na_transforms.cpp -o
build/temp.aix-5.3-2.5/src/_na_transforms.o -DNUMARRAY=3D1
"./CXX/Objects.hxx", line 1436.100: 1540-0140 (S) The text "&" is
unexpected.  "SeqBase<T>::iterator" may be undeclared, ambiguous, or may
require "typename" qualification.

I've got libstd++.a and libstdc++.la in
/Produits/publics/powerpc.AIX.5.3/gcc/3.1/include


Setting CXX to g++ doesn't help :
g++ cc_r -bI:/usr/local/lib/python2.5/config/python.exp
build/temp.aix-5.3-2.5/src/agg.o
build/temp.aix-5.3-2.5/agg23/src/agg_trans_affine.o
build/temp.aix-5.3-2.5/agg23/src/agg_path_storage.o
build/temp.aix-5.3-2.5/agg23/src/agg_bezier_arc.o
build/temp.aix-5.3-2.5/agg23/src/agg_vcgen_dash.o
build/temp.aix-5.3-2.5/agg23/src/agg_vcgen_stroke.o
build/temp.aix-5.3-2.5/agg23/src/agg_rasterizer_scanline_aa.o
build/temp.aix-5.3-2.5/agg23/src/agg_curves.o -lstdc++ -lm -o
build/lib.aix-5.3-2.5/matplotlib/_agg.so
g++: cc_r: No such file or directory
error: command 'g++' failed with exit status 1

Is there any other way to select the c++ compiler? Because setting the
CXX variable doesn't seem to work.

belinda thom wrote:
> Thanks Eric!
> 
> I'm kind of afraid to upgrade to the SVN version b/c I fear something 
> else might break (I've yet to install matplotlib from source b/c of a 
> few nagging Mac OS X issues).
> 
> What would you recommend?

I don't know; I see from your earlier message that Tim Leslie says some 
of the things you are trying to do are still broken, and his patches are 
what John committed, so svn is certainly no better than what Tim is 
working with now.

I don't think that anything has changed since 0.90 in such a way as to 
require changing both the compiled parts and the pure python parts, so 
you could try simply replacing the pure python part.

> 
> There's also the issue of having to upgrade student machines for which 
> I'm not admin. Grrr.

That certainly makes the prospect of doing any kind of upgrade less 
attractive.  I don't have a good solution.
> 
> --b
> 
> On Apr 11, 2007, at 6:19 PM, Eric Firing wrote:
> 
>> Belinda,
>>
>> John has checked in some fixes for axes3d recently, and your test code 
>> now runs with the version in svn--except for the problem that you 
>> should use P.show() only once in a given script.  (I think there is 
>> some backend for which you can get away with using it more than once, 
>> but this is not good practice.)
>>
>> Eric
>>
>> belinda thom wrote:
>>> On Apr 11, 2007, at 4:47 PM, belinda thom wrote:
>>>> Hi,
>>>>
>>>> I'm replying to this older thread
>>>>
>>>> http://www.mail-archive.com/mat...@li.../ 
>>>> msg02100.html
>>>>
>>>> because it relates very much to my recent problem, posted at http:// 
>>>> www.mail-archive.com/mat...@li.../ 
>>>> msg03037.html. This problem seems more general than just the two  
>>>> circumstances you mentioned.
>>>>
>>>> I'm finding the problem w/any use plot3D in matplotlib.axes3d, e.g.:
>>> FYI: It also is a problem when using plot3d
>>> --b
>>> ------------------------------------------------------------------------- 
>>>
>>> Take Surveys. Earn Cash. Influence the Future of IT
>>> Join SourceForge.net's Techsay panel and you'll get the chance to 
>>> share your
>>> opinions on IT & business topics through brief surveys-and earn cash
>>> http://www.techsay.com/default.php?page=join.php&p=sourceforge&CID=DEVDEV 
>>>
>>> _______________________________________________
>>> Matplotlib-users mailing list
>>> Mat...@li...
>>> https://lists.sourceforge.net/lists/listinfo/matplotlib-users
>>
> 

Thanks Eric!

I'm kind of afraid to upgrade to the SVN version b/c I fear something  
else might break (I've yet to install matplotlib from source b/c of a  
few nagging Mac OS X issues).

What would you recommend?

There's also the issue of having to upgrade student machines for  
which I'm not admin. Grrr.

--b

On Apr 11, 2007, at 6:19 PM, Eric Firing wrote:

> Belinda,
>
> John has checked in some fixes for axes3d recently, and your test  
> code now runs with the version in svn--except for the problem that  
> you should use P.show() only once in a given script.  (I think  
> there is some backend for which you can get away with using it more  
> than once, but this is not good practice.)
>
> Eric
>
> belinda thom wrote:
>> On Apr 11, 2007, at 4:47 PM, belinda thom wrote:
>>> Hi,
>>>
>>> I'm replying to this older thread
>>>
>>> http://www.mail-archive.com/matplotlib- 
>>> us...@li.../ msg02100.html
>>>
>>> because it relates very much to my recent problem, posted at  
>>> http:// www.mail-archive.com/matplotlib- 
>>> us...@li.../ msg03037.html. This problem seems  
>>> more general than just the two  circumstances you mentioned.
>>>
>>> I'm finding the problem w/any use plot3D in matplotlib.axes3d, e.g.:
>> FYI: It also is a problem when using plot3d
>> --b
>> --------------------------------------------------------------------- 
>> ----
>> Take Surveys. Earn Cash. Influence the Future of IT
>> Join SourceForge.net's Techsay panel and you'll get the chance to  
>> share your
>> opinions on IT & business topics through brief surveys-and earn cash
>> http://www.techsay.com/default.php? 
>> page=join.php&p=sourceforge&CID=DEVDEV
>> _______________________________________________
>> Matplotlib-users mailing list
>> Mat...@li...
>> https://lists.sourceforge.net/lists/listinfo/matplotlib-users
>

Forwarding this to the list....

I'm curious if you run into another problem I recently reported. Can  
you do, e.g.:

import copy
import pylab as P
import matplotlib.axes3d as P3
def test() :
     [X,Y] = P.meshgrid(P.linspace(-3,3,7),P.linspace(-3,3,7))
     Z = copy.deepcopy(X)
     Z1 = copy.deepcopy(X)
     for i in xrange(len(X)) :
         for j in xrange(len(X[0])) :
             Z[i][j] = X[i][j]*Y[i][j]
             Z1[i][j] = X[i][j]*Y[i][j] - 5
     P.close('all')
     fig = P.figure()
     ax = P3.Axes3D(fig)
     ax.plot3d(ravel(X),ravel(Y),ravel(Z),'r.-')

or do you instead get an error in matplotlib's axes.py, complaining  
about self.autoscale_view and a scalex parameter? For more on the  
hack I made to axes.py file which then allowed me to use plot3d, see  
http://www.mail-archive.com/mat...@li.../ 
msg03041.html.

Thanks for providing  more info.

--b

Begin forwarded message:

> From: "Richard Brown" <rg...@gm...>
> Date: April 11, 2007 8:41:37 PM PDT
> To: "belinda thom" <bt...@cs...>
> Subject: Re: [Matplotlib-users] 3d plotting question
>
> On 12/04/07, belinda thom <bt...@cs...> wrote:
>> I also seem to have some other 3D plotting problems. Again, following
>> some of the demo/test advice in the cookbook, I tried things like
>>
>>    import pylab as p
>>    import matplotlib.axes3d as P3
>>    P3.test_surface()
>
> I too just encountered the same problem - what is the last version of
> mpl for which doing this worked? (I'm using 0.9 revision 3131)
>
> -- 
> Richard Brown
> Ph.D. Candidate
> Dept. of Mechanical Engineering
> University of Canterbury, NZ

> Hi Belinda,
>
> I've been playing with 3D plots and scatter plots in the past few days
> and I've been able to get them working. You should be able to pass in
> a c=<color> parameter as you would for a normal 2d scatter plot. I've
> been doing this and it's working for me.

Interesting.

I'm on Mac OS X 4.8.9, running Python 2.4.4, using Matplotlib 0.87.7  
w/TkAgg (I might be using Wx, but would have to get a problem w/my  
machine's and all the students' lab machines wx libraries updated).

When I try using the c='r' parameter, I run into problems. I also  
tried c='red'. One big problem for me is that which kwargs can be  
used where and what values they can take on is not well documented.

For instance, in plot3d (as opposed to plot3D) you can pass the  
typical Matplotlib 'r.-' string and it just works. But in  
plot_wireframe, for example, it appears you must use different kwargs  
(for instance, linestyle='solid' and color='<some RGB value>'). I am  
not sure about these notions, b/c I've had to deduce them from the  
code and various google-based snippets.

That's why I'm asking for more guidance about the various kwarg usages.

Is there any good place to find more cohesive info online? The most  
useful place I could find seemed to be in matplotlib's  
collections.py, an __init__ function's documentation (that's where I  
deduced, for instance, the linestyle and RGB color info). If there  
isn't such a place, someone should probably modify the matplotlib 3d  
cookbook. I'd be willing to do that but I'd first need to have a  
better understanding about the various functions' kwargs :-).

I'll append a case I used to generate an error, as well as the error  
reporting below. I'm also not quite sure how to use the errors that  
are reported to help me track down where the problems might be, so if  
anyone has some advice on that, it might help me better help myself  
in the future.

As always, thx,

--b


Example Code:
-------------
import copy
import pylab as P
import matplotlib.axes3d as P3
def test() :
     [X,Y] = P.meshgrid(P.linspace(-3,3,7),P.linspace(-3,3,7))
     Z = copy.deepcopy(X)
     Z1 = copy.deepcopy(X)
     for i in xrange(len(X)) :
         for j in xrange(len(X[0])) :
             Z[i][j] = X[i][j]*Y[i][j]
             Z1[i][j] = X[i][j]*Y[i][j] - 5
     P.close('all')
     fig = P.figure()
     ax = P3.Axes3D(fig)
     ax.Plot3D(ravel(X),ravel(Y),ravel(Z),c='r')

In [213]: T.test()
------------------------------------------------------------------------ 
---
exceptions.AttributeError                            Traceback (most  
recent call last)

/Users/bthom/belinda/mills/aicourse/material/week12/lec/examples/ 
<ipython console>

/Users/bthom/belinda/mills/aicourse/material/week12/lec/examples/ 
threeDa.py in test()
      11             Z1[i][j] = X[i][j]*Y[i][j] - 5
      12     P.close('all')
      13     fig = P.figure()
      14     ax = P3.Axes3D(fig)
---> 15     ax.Plot3D(ravel(X),ravel(Y),ravel(Z),c='red')

/Library/Frameworks/Python.framework/Versions/2.4/lib/python2.4/site- 
packages/matplotlib/axes3d.py in __getattr__(self, k)
     660
     661     def __getattr__(self, k):
--> 662         return getattr(self.wrapped,k)
     663
     664     def __setattr__(self, k,v):

AttributeError: Axes3DI instance has no attribute 'Plot3D'
 > /Library/Frameworks/Python.framework/Versions/2.4/lib/python2.4/ 
site-packages/matplotlib/axes3d.py(662)__getattr__()
     661     def __getattr__(self, k):
--> 662         return getattr(self.wrapped,k)
     663

ipdb> 

Thanks for the input.

Its easy for me to patch my own machine, but for students in my class  
who are using lab machines, its more difficult. Hopefully I can get  
someone to upgrade the machines in the lab I'm using. (Its difficult  
to get facilities people to agree to update coursework installs mid- 
semester).

Morale of my story: don't expect 3d plotting to necessarily work / be  
easy if you don't carefully test it first :-(.

I too would be interested in robustifying / making more accessible  
the 3d stuff, but will have to wait until after the class I'm  
teaching ends.

--b


On Apr 11, 2007, at 6:57 PM, Tim Leslie wrote:

> Even in the lastest svn version these test methods are all broken to a
> greater or lesser degree (as I found out a couple of days ago). If I
> find time over the weekend I'll try to fix them up. In the mean time,
> if you change them to look something like
>
> fig = pylab.figure()
> ax = Axes3d(fig)
> ... etc
>
> it should get you one step closer to having them working. I'll post
> back here if/when I manage to get them cleaned up.
>
> Cheers,
>
> Tim
>

Belinda,

John has checked in some fixes for axes3d recently, and your test code 
now runs with the version in svn--except for the problem that you should 
use P.show() only once in a given script.  (I think there is some 
backend for which you can get away with using it more than once, but 
this is not good practice.)

Eric

belinda thom wrote:
> On Apr 11, 2007, at 4:47 PM, belinda thom wrote:
> 
>> Hi,
>>
>> I'm replying to this older thread
>>
>> http://www.mail-archive.com/mat...@li.../ 
>> msg02100.html
>>
>> because it relates very much to my recent problem, posted at http:// 
>> www.mail-archive.com/mat...@li.../ 
>> msg03037.html. This problem seems more general than just the two  
>> circumstances you mentioned.
>>
>> I'm finding the problem w/any use plot3D in matplotlib.axes3d, e.g.:
> 
> FYI: It also is a problem when using plot3d
> 
> --b
> 
> 
> -------------------------------------------------------------------------
> Take Surveys. Earn Cash. Influence the Future of IT
> Join SourceForge.net's Techsay panel and you'll get the chance to share your
> opinions on IT & business topics through brief surveys-and earn cash
> http://www.techsay.com/default.php?page=join.php&p=sourceforge&CID=DEVDEV
> _______________________________________________
> Matplotlib-users mailing list
> Mat...@li...
> https://lists.sourceforge.net/lists/listinfo/matplotlib-users