matplotlib-users Mailing List for matplotlib

Can matplotlib be used in scripts with multiple processes, such as
when using the multiprocessing module?
I had some difficulty getting code to work in which I was carrying out
some long computations and wanted to show intermediate results in a
separate process so as not to interrupt the main computation. The
problem distilled down to trying to use the multiprocessing module and
matplotlib together. This is a code snippet which I expected to work,
but doesn't. Is this a matplotlib issue, some intricate issue with
GUIs and processes, or something with my understanding of the subject?
Would you expect this code to work? Is there any easy workaround?
from pylab import *
import multiprocessing
class myclass(object):
 def plotter(self):
 plot([1,2,3,4])
 show()		
 def mainfunction(self):
 a = arange(1, 10, 0.1)
 newprocess = multiprocessing.Process(target=self.plotter)
 newprocess.start()
 newprocess.join()
 plot(a, sin(a))
 show()
if __name__== '__main__':
 a = myclass()	
 a.mainfunction()
Thanks,
Sameer

Hi,
I am trying to change the format of the tick labels in a loglog plot from
exponential to decimal.
Right now the labels are 10^2 10 1 10^-1 10^-2
and I would like them to be 100 10 1 0.1 0.01
I tried finding easy explanations in the web with no luck. By the way I have
no clue what a class is
and the word instantiation makes me crossed eyed
If any charitable soul would help me I would greatly appreciate it
gino

> On 2010年10月31日, David Kremer wrote:
> > File
> > "/usr/lib/python2.7/site-packages/matplotlib/backends/backend_agg.py",
> 
> python 2.7 ^^
> 
> > RuntimeError: Could not open facefile /usr/lib/python2.6/site-
> > packages/matplotlib/mpl-data/fonts/ttf/Vera.ttf; Cannot_Open_Resource
> 
> python 2.6 ^^
> 
> I believe this problem is fixed in the next matplotlib release. A
> workaround is to remove a local font cache file:
> 
> rm ~/.matplotlib/fontList.cache
> 
> HTH,
> daryl
It worked. Thank you.
I also did a symbolic link, but your method worked, so I deleted it.
Thank you.
David

On 2010年10月31日, David Kremer wrote:
> File "/usr/lib/python2.7/site-packages/matplotlib/backends/backend_agg.py",
python 2.7 ^^
> RuntimeError: Could not open facefile /usr/lib/python2.6/site-
> packages/matplotlib/mpl-data/fonts/ttf/Vera.ttf; Cannot_Open_Resource
python 2.6 ^^
I believe this problem is fixed in the next matplotlib release. A 
workaround is to remove a local font cache file:
rm ~/.matplotlib/fontList.cache
HTH,
daryl

Hello, Im running the last matplotlib version.
I just want to plot an histogram. Here is the complete backtrace, complaining 
about the Vera.ttf font missing :
Any idea ?
Traceback (most recent call last):
 File "/usr/lib/python2.7/site-packages/matplotlib/backends/backend_gtk.py", 
line 393, in expose_event
 self._render_figure(self._pixmap, w, h)
 File "/usr/lib/python2.7/site-packages/matplotlib/backends/backend_gtkagg.py", 
line 75, in _render_figure
 FigureCanvasAgg.draw(self)
 File "/usr/lib/python2.7/site-packages/matplotlib/backends/backend_agg.py", 
line 394, in draw
 self.figure.draw(self.renderer)
 File "/usr/lib/python2.7/site-packages/matplotlib/artist.py", line 55, in 
draw_wrapper
 draw(artist, renderer, *args, **kwargs)
 File "/usr/lib/python2.7/site-packages/matplotlib/figure.py", line 798, in draw
 func(*args)
 File "/usr/lib/python2.7/site-packages/matplotlib/artist.py", line 55, in 
draw_wrapper
 draw(artist, renderer, *args, **kwargs)
 File "/usr/lib/python2.7/site-packages/matplotlib/axes.py", line 1934, in draw
 a.draw(renderer)
 File "/usr/lib/python2.7/site-packages/matplotlib/artist.py", line 55, in 
draw_wrapper
 draw(artist, renderer, *args, **kwargs)
 File "/usr/lib/python2.7/site-packages/matplotlib/axis.py", line 1017, in draw
 tick.draw(renderer)
 File "/usr/lib/python2.7/site-packages/matplotlib/artist.py", line 55, in 
draw_wrapper
 draw(artist, renderer, *args, **kwargs)
 File "/usr/lib/python2.7/site-packages/matplotlib/axis.py", line 234, in draw
 self.label1.draw(renderer)
 File "/usr/lib/python2.7/site-packages/matplotlib/artist.py", line 55, in 
draw_wrapper
 draw(artist, renderer, *args, **kwargs)
 File "/usr/lib/python2.7/site-packages/matplotlib/text.py", line 524, in draw
 bbox, info = self._get_layout(renderer)
 File "/usr/lib/python2.7/site-packages/matplotlib/text.py", line 298, in 
_get_layout
 ismath=False)
 File "/usr/lib/python2.7/site-packages/matplotlib/backends/backend_agg.py", 
line 180, in get_text_width_height_descent
 font = self._get_agg_font(prop)
 File "/usr/lib/python2.7/site-packages/matplotlib/backends/backend_agg.py", 
line 221, in _get_agg_font
 font = FT2Font(str(fname))
RuntimeError: Could not open facefile /usr/lib/python2.6/site-
packages/matplotlib/mpl-data/fonts/ttf/Vera.ttf; Cannot_Open_Resource
Traceback (most recent call last):
 File "/usr/lib/python2.7/site-packages/matplotlib/backends/backend_gtk.py", 
line 393, in expose_event
 self._render_figure(self._pixmap, w, h)
 File "/usr/lib/python2.7/site-packages/matplotlib/backends/backend_gtkagg.py", 
line 75, in _render_figure
 FigureCanvasAgg.draw(self)
 File "/usr/lib/python2.7/site-packages/matplotlib/backends/backend_agg.py", 
line 394, in draw
 self.figure.draw(self.renderer)
 File "/usr/lib/python2.7/site-packages/matplotlib/artist.py", line 55, in 
draw_wrapper
 draw(artist, renderer, *args, **kwargs)
 File "/usr/lib/python2.7/site-packages/matplotlib/figure.py", line 798, in draw
 func(*args)
 File "/usr/lib/python2.7/site-packages/matplotlib/artist.py", line 55, in 
draw_wrapper
 draw(artist, renderer, *args, **kwargs)
 File "/usr/lib/python2.7/site-packages/matplotlib/axes.py", line 1934, in draw
 a.draw(renderer)
 File "/usr/lib/python2.7/site-packages/matplotlib/artist.py", line 55, in 
draw_wrapper
 draw(artist, renderer, *args, **kwargs)
 File "/usr/lib/python2.7/site-packages/matplotlib/axis.py", line 1017, in draw
 tick.draw(renderer)
 File "/usr/lib/python2.7/site-packages/matplotlib/artist.py", line 55, in 
draw_wrapper
 draw(artist, renderer, *args, **kwargs)
 File "/usr/lib/python2.7/site-packages/matplotlib/axis.py", line 234, in draw
 self.label1.draw(renderer)
 File "/usr/lib/python2.7/site-packages/matplotlib/artist.py", line 55, in 
draw_wrapper
 draw(artist, renderer, *args, **kwargs)
 File "/usr/lib/python2.7/site-packages/matplotlib/text.py", line 524, in draw
 bbox, info = self._get_layout(renderer)
 File "/usr/lib/python2.7/site-packages/matplotlib/text.py", line 298, in 
_get_layout
 ismath=False)
 File "/usr/lib/python2.7/site-packages/matplotlib/backends/backend_agg.py", 
line 180, in get_text_width_height_descent
 font = self._get_agg_font(prop)
 File "/usr/lib/python2.7/site-packages/matplotlib/backends/backend_agg.py", 
line 221, in _get_agg_font
 font = FT2Font(str(fname))
RuntimeError: Could not open facefile /usr/lib/python2.6/site-
packages/matplotlib/mpl-data/fonts/ttf/Vera.ttf; Cannot_Open_Resource

On Wed, Oct 27, 2010 at 10:36 PM, Matthieu Huin
<mat...@wa...> wrote:
> Anyone knows of a comprehensive doc or tutorial on that subject ?
Unfortunately there is no such things as far as I know.
The agg_filter itself has relatively simple api (from the
documentation of backend_agg.stop_filter).
 """
 Save the plot in the current canvas as a image and apply
 the *post_processing* function.
 def post_processing(image, dpi):
 # ny, nx, depth = image.shape
 # image (numpy array) has RGBA channels and has a depth of 4.
 ...
 # create a new_image (numpy array of 4 channels, size can be
 # different). The resulting image may have offsets from
 # lower-left corner of the original image
 return new_image, offset_x, offset_y
 The saved renderer is restored and the returned image from
 post_processing is plotted (using draw_image) on it.
 """
So, any callable object with above api can be used as a filter. And
the classes in the above example describes just one way to create such
filters.
The agg_filter functionality is a very experimental feature. While I
contributed the example code, I myself rarely use this feature. And
I'm afraid that I may not be able to find enough time to maintain
and/or improve the code.
Regards,
-JJ

On 10/31/10 5:44 AM, John wrote:
> Jeff,
>
> I just built netcdf with the following script:
>
> #!/bin/bash
>
> ./configure --enable-netcdf-4 --enable-hdf4 --enable-hdf4-file-tests
> --enable-pnetcdf --enable-cxx-4 --with-hdf5=/flex_wrk/lib64
> --enable-shared --prefix=/flex_wrk/lib64
> make
> make install
John: You forget --with-hdf4=<where hdf4 is installed>.
-Jeff
>
> Everything seemed to work fine.
>
> I then rebuilt netcdf4-python and got no errors.
> IE:
> export HDF5_DIR=/flex_wrk/lib64
> export NETCDF4_DIR=/flex_wrk/lib64
> python setup.py build
> python setup.py install --prefix=/custom...
>
> However, when I try to read a hdf4 file in Ipython I get:
> In [1]: from netCDF4 import Dataset as NetCDFFile
>
> In [2]: nci = NetCDFFile('scia_200910.hdf')
> ---------------------------------------------------------------------------
> RuntimeError Traceback (most recent call last)
>
> /xnilu_wrk/flex_wrk/EMISSIONS/<ipython console> in<module>()
>
> /xnilu_wrk/flex_wrk/bin64/site-packages/netCDF4.so in
> netCDF4.Dataset.__init__ (netCDF4.c:6336)()
>
> RuntimeError: Attempt to use feature that was not turned on when
> netCDF was built.
>
>
> What is it that I need to do to 'build' netCDF and netcdf4-python with
> the support?
>
> Thank you,
> john
>
>
> On Sun, Oct 31, 2010 at 2:53 AM, Jeff Whitaker<js...@fa...> wrote:
>> On 10/30/10 5:05 PM, John wrote:
>>> Thank you. I didn't think they used it, I just wanted to know if it
>>> was safe to do what you suggested to try. It does seem to work.
>>>
>>> Another question, can netcd4-python READ HDF files? Or is it just that
>>> it creates files that are readable? If it's the latter, does anyone
>>> have a suggestion for a python module that can read both hdf4 and
>>> hdf5??
>>>
>>> Thank you,
>>> john
>>>
>>>
>>> PS: Jeff, this module seems to work very well! And I am very thankful
>>> for the introduction of compression.
>> John: If you enabled hdf4 support when you built the netcdf lib, it will
>> read hdf4 files in 'scientific dataset' files.
>>
>> -Jeff
>>>
>>> On Sat, Oct 30, 2010 at 2:55 PM, Jeff Whitaker<js...@fa...> wrote:
>>>> On 10/30/10 3:41 AM, John wrote:
>>>>> I've searched through the netcdf4-python trunk, but I find no
>>>>> reference of NetCDFFile as in how the basemap or Scientific.IO.NetCDF
>>>>> modules use it.
>>>>>
>>>>> For compatibility with my old scripts, is it safe to simply:
>>>>>
>>>>> from netCDF4 import Dataset as NetCDFFile
>>>>>
>>>>> ??
>>>>>
>>>>> Thanks,
>>>>> john
>>>>>
>>>> John: Neither basemap or Scientific.IO.NetCDF uses it. What gave you
>>>> the impression they did? There is a 'NetCDFFile' in function in
>>>> basemap, but it is based on the pure-python netCDF module pupynere. It
>>>> doesn't support netcdf-4 and I only recommend using it for the
>>>> occasional light use.
>>>>
>>>> The netcdf4-python API is mostly compatible with Scientific.IO.NetCDF,
>>>> so if you have old script that use the latter, you can try:
>>>>
>>>> from netCDF4 import Dataset as NetCDFFile
>>>>
>>>> The docs for netcdf4-python are at
>>>>
>>>> http://netcdf4-python.googlecode.com/svn/trunk/docs/netCDF4-module.html
>>>>
>>>>
>>>> -Jeff
>>>>
>>>>
>>>>
>>>> ------------------------------------------------------------------------------
>>>> Nokia and AT&T present the 2010 Calling All Innovators-North America
>>>> contest
>>>> Create new apps& games for the Nokia N8 for consumers in U.S. and
>>>> Canada
>>>> 10ドル million total in prizes - 4ドルM cash, 500 devices, nearly 6ドルM in
>>>> marketing
>>>> Develop with Nokia Qt SDK, Web Runtime, or Java and Publish to Ovi Store
>>>> http://p.sf.net/sfu/nokia-dev2dev
>>>> _______________________________________________
>>>> Matplotlib-users mailing list
>>>> Mat...@li...
>>>> https://lists.sourceforge.net/lists/listinfo/matplotlib-users
>>>>
>>>
>>
>
>

Jeff,
I just built netcdf with the following script:
#!/bin/bash
./configure --enable-netcdf-4 --enable-hdf4 --enable-hdf4-file-tests
--enable-pnetcdf --enable-cxx-4 --with-hdf5=/flex_wrk/lib64
--enable-shared --prefix=/flex_wrk/lib64
make
make install
Everything seemed to work fine.
I then rebuilt netcdf4-python and got no errors.
IE:
export HDF5_DIR=/flex_wrk/lib64
export NETCDF4_DIR=/flex_wrk/lib64
python setup.py build
python setup.py install --prefix=/custom...
However, when I try to read a hdf4 file in Ipython I get:
In [1]: from netCDF4 import Dataset as NetCDFFile
In [2]: nci = NetCDFFile('scia_200910.hdf')
---------------------------------------------------------------------------
RuntimeError Traceback (most recent call last)
/xnilu_wrk/flex_wrk/EMISSIONS/<ipython console> in <module>()
/xnilu_wrk/flex_wrk/bin64/site-packages/netCDF4.so in
netCDF4.Dataset.__init__ (netCDF4.c:6336)()
RuntimeError: Attempt to use feature that was not turned on when
netCDF was built.
What is it that I need to do to 'build' netCDF and netcdf4-python with
the support?
Thank you,
john
On Sun, Oct 31, 2010 at 2:53 AM, Jeff Whitaker <js...@fa...> wrote:
> On 10/30/10 5:05 PM, John wrote:
>>
>> Thank you. I didn't think they used it, I just wanted to know if it
>> was safe to do what you suggested to try. It does seem to work.
>>
>> Another question, can netcd4-python READ HDF files? Or is it just that
>> it creates files that are readable? If it's the latter, does anyone
>> have a suggestion for a python module that can read both hdf4 and
>> hdf5??
>>
>> Thank you,
>> john
>>
>>
>> PS: Jeff, this module seems to work very well! And I am very thankful
>> for the introduction of compression.
>
> John: If you enabled hdf4 support when you built the netcdf lib, it will
> read hdf4 files in 'scientific dataset' files.
>
> -Jeff
>>
>>
>> On Sat, Oct 30, 2010 at 2:55 PM, Jeff Whitaker<js...@fa...> wrote:
>>>
>>> On 10/30/10 3:41 AM, John wrote:
>>>>
>>>> I've searched through the netcdf4-python trunk, but I find no
>>>> reference of NetCDFFile as in how the basemap or Scientific.IO.NetCDF
>>>> modules use it.
>>>>
>>>> For compatibility with my old scripts, is it safe to simply:
>>>>
>>>> from netCDF4 import Dataset as NetCDFFile
>>>>
>>>> ??
>>>>
>>>> Thanks,
>>>> john
>>>>
>>> John: Neither basemap or Scientific.IO.NetCDF uses it. What gave you
>>> the impression they did? There is a 'NetCDFFile' in function in
>>> basemap, but it is based on the pure-python netCDF module pupynere. It
>>> doesn't support netcdf-4 and I only recommend using it for the
>>> occasional light use.
>>>
>>> The netcdf4-python API is mostly compatible with Scientific.IO.NetCDF,
>>> so if you have old script that use the latter, you can try:
>>>
>>> from netCDF4 import Dataset as NetCDFFile
>>>
>>> The docs for netcdf4-python are at
>>>
>>> http://netcdf4-python.googlecode.com/svn/trunk/docs/netCDF4-module.html
>>>
>>>
>>> -Jeff
>>>
>>>
>>>
>>> ------------------------------------------------------------------------------
>>> Nokia and AT&T present the 2010 Calling All Innovators-North America
>>> contest
>>> Create new apps& games for the Nokia N8 for consumers in U.S. and
>>> Canada
>>> 10ドル million total in prizes - 4ドルM cash, 500 devices, nearly 6ドルM in
>>> marketing
>>> Develop with Nokia Qt SDK, Web Runtime, or Java and Publish to Ovi Store
>>> http://p.sf.net/sfu/nokia-dev2dev
>>> _______________________________________________
>>> Matplotlib-users mailing list
>>> Mat...@li...
>>> https://lists.sourceforge.net/lists/listinfo/matplotlib-users
>>>
>>
>>
>
>
-- 
Configuration
``````````````````````````
Plone 2.5.3-final,
CMF-1.6.4,
Zope (Zope 2.9.7-final, python 2.4.4, linux2),
Python 2.6
PIL 1.1.6
Mailman 2.1.9
Postfix 2.4.5
Procmail v3.22 2001年09月10日
Basemap: 1.0
Matplotlib: 1.0.0

On 10/30/10 5:05 PM, John wrote:
> Thank you. I didn't think they used it, I just wanted to know if it
> was safe to do what you suggested to try. It does seem to work.
>
> Another question, can netcd4-python READ HDF files? Or is it just that
> it creates files that are readable? If it's the latter, does anyone
> have a suggestion for a python module that can read both hdf4 and
> hdf5??
>
> Thank you,
> john
>
>
> PS: Jeff, this module seems to work very well! And I am very thankful
> for the introduction of compression.
John: If you enabled hdf4 support when you built the netcdf lib, it will 
read hdf4 files in 'scientific dataset' files.
-Jeff
>
>
> On Sat, Oct 30, 2010 at 2:55 PM, Jeff Whitaker<js...@fa...> wrote:
>> On 10/30/10 3:41 AM, John wrote:
>>> I've searched through the netcdf4-python trunk, but I find no
>>> reference of NetCDFFile as in how the basemap or Scientific.IO.NetCDF
>>> modules use it.
>>>
>>> For compatibility with my old scripts, is it safe to simply:
>>>
>>> from netCDF4 import Dataset as NetCDFFile
>>>
>>> ??
>>>
>>> Thanks,
>>> john
>>>
>> John: Neither basemap or Scientific.IO.NetCDF uses it. What gave you
>> the impression they did? There is a 'NetCDFFile' in function in
>> basemap, but it is based on the pure-python netCDF module pupynere. It
>> doesn't support netcdf-4 and I only recommend using it for the
>> occasional light use.
>>
>> The netcdf4-python API is mostly compatible with Scientific.IO.NetCDF,
>> so if you have old script that use the latter, you can try:
>>
>> from netCDF4 import Dataset as NetCDFFile
>>
>> The docs for netcdf4-python are at
>>
>> http://netcdf4-python.googlecode.com/svn/trunk/docs/netCDF4-module.html
>>
>>
>> -Jeff
>>
>>
>> ------------------------------------------------------------------------------
>> Nokia and AT&T present the 2010 Calling All Innovators-North America contest
>> Create new apps& games for the Nokia N8 for consumers in U.S. and Canada
>> 10ドル million total in prizes - 4ドルM cash, 500 devices, nearly 6ドルM in marketing
>> Develop with Nokia Qt SDK, Web Runtime, or Java and Publish to Ovi Store
>> http://p.sf.net/sfu/nokia-dev2dev
>> _______________________________________________
>> Matplotlib-users mailing list
>> Mat...@li...
>> https://lists.sourceforge.net/lists/listinfo/matplotlib-users
>>
>
>

Thank you. I didn't think they used it, I just wanted to know if it
was safe to do what you suggested to try. It does seem to work.
Another question, can netcd4-python READ HDF files? Or is it just that
it creates files that are readable? If it's the latter, does anyone
have a suggestion for a python module that can read both hdf4 and
hdf5??
Thank you,
john
PS: Jeff, this module seems to work very well! And I am very thankful
for the introduction of compression.
On Sat, Oct 30, 2010 at 2:55 PM, Jeff Whitaker <js...@fa...> wrote:
> On 10/30/10 3:41 AM, John wrote:
>> I've searched through the netcdf4-python trunk, but I find no
>> reference of NetCDFFile as in how the basemap or Scientific.IO.NetCDF
>> modules use it.
>>
>> For compatibility with my old scripts, is it safe to simply:
>>
>> from netCDF4 import Dataset as NetCDFFile
>>
>> ??
>>
>> Thanks,
>> john
>>
> John: Neither basemap or Scientific.IO.NetCDF uses it. What gave you
> the impression they did? There is a 'NetCDFFile' in function in
> basemap, but it is based on the pure-python netCDF module pupynere. It
> doesn't support netcdf-4 and I only recommend using it for the
> occasional light use.
>
> The netcdf4-python API is mostly compatible with Scientific.IO.NetCDF,
> so if you have old script that use the latter, you can try:
>
> from netCDF4 import Dataset as NetCDFFile
>
> The docs for netcdf4-python are at
>
> http://netcdf4-python.googlecode.com/svn/trunk/docs/netCDF4-module.html
>
>
> -Jeff
>
>
> ------------------------------------------------------------------------------
> Nokia and AT&T present the 2010 Calling All Innovators-North America contest
> Create new apps & games for the Nokia N8 for consumers in U.S. and Canada
> 10ドル million total in prizes - 4ドルM cash, 500 devices, nearly 6ドルM in marketing
> Develop with Nokia Qt SDK, Web Runtime, or Java and Publish to Ovi Store
> http://p.sf.net/sfu/nokia-dev2dev
> _______________________________________________
> Matplotlib-users mailing list
> Mat...@li...
> https://lists.sourceforge.net/lists/listinfo/matplotlib-users
>
-- 
Configuration
``````````````````````````
Plone 2.5.3-final,
CMF-1.6.4,
Zope (Zope 2.9.7-final, python 2.4.4, linux2),
Python 2.6
PIL 1.1.6
Mailman 2.1.9
Postfix 2.4.5
Procmail v3.22 2001年09月10日
Basemap: 1.0
Matplotlib: 1.0.0

Hello,
I am trying to set tick labels for a matshow plot. This works just fine
if I only work on the X axis:
 res = np.diag(np.arange(10)) # dummy data
 modes = [ (x+1, 0) for x in range(5) ] # dummy data
 
 fig = plt.figure()
 ax = fig.add_subplot(111)
 cs = ax.matshow(res)
 fig.colorbar(cs, ticks=np.arange(0,-10,-1), shrink=0.8)
 ax.set_xticklabels(['%d/%d' % (x[1], x[0]) for x in modes])
 ax.set_xticks(2*np.arange(len(modes))+0.5)
but if I try to do the same think with the Y axis, everything looks
messed up (e.g. the matrix is no longer square):
 ax.set_yticklabels(['%d/%d' % (x[1], x[0]) for x in modes])
 ax.set_yticks(2*np.arange(len(modes))+0.5)
 
What am I doing wrong?
Thanks,
 -Nikolaus
-- 
 »Time flies like an arrow, fruit flies like a Banana.«
 PGP fingerprint: 5B93 61F8 4EA2 E279 ABF6 02CF A9AD B7F8 AE4E 425C

lol.
Finally found the magic parameter. borderaxespad.
import pylab,matplotlib
pylab.plot([1,2,3,4,5],[1,2,3,4,5],"bo",label="data")
ax = pylab.gca()
pylab.legend(bbox_to_anchor=(0,1),loc=2,borderaxespad=0)
# draw an X to make sure of coordinates.
ax.lines.append(matplotlib.lines.Line2D((0,1),(0,1),transform=ax.transAxes,color='blue'))
ax.lines.append(matplotlib.lines.Line2D((0,1),(1,0),transform=ax.transAxes,color='red'))
pylab.show()
I don't know how I missed that parameter after reading the docs 100 times. :(
Sorry for bothering the list with something so trivial. I had a major
"can't see the forest through the trees" moment.
-- 
Daniel Hyams
dh...@gm...

With Paul's suggestions (greatly appreciated!) I almost have this
thing working....there is still something strange going on in that
when I call get_window_extent() on the legend, I always get ones and
zeros no matter where it is...but that's for another post on another
day.
This one, I hope, is an easy one. The last thing I need to do is
position the legend precisely. For example, the code below doesn't
quite position the top left corner of the legend at the top left
corner in the plot:
import pylab,matplotlib
pylab.plot([1,2,3,4,5],[1,2,3,4,5],"bo",label="data")
ax = pylab.gca()
pylab.legend(bbox_to_anchor=(0,1),loc=2)
# draw an X to make sure of coordinates.
ax.lines.append(matplotlib.lines.Line2D((0,1),(0,1),transform=ax.transAxes,color='blue'))
ax.lines.append(matplotlib.lines.Line2D((0,1),(1,0),transform=ax.transAxes,color='red'))
pylab.show()
It's close, but not there.....so the question becomes, how do I place
a legend exactly at the coordinates that I want?
On Sat, Oct 30, 2010 at 1:16 AM, Paul Ivanov <piv...@gm...> wrote:
> Daniel Hyams, on 2010年10月29日 23:48, wrote:
>> Thanks Paul! Your suggestion got me part of the way, but I've run
>> into another problem...I'm using draggable legends, I'm also wanting
>> to fetch the current position of the legend after a drag. The
>> draggable legend always updates 'loc', and not 'bbox_to_anchor', so
>> I'm afraid that I'm stuck manipulating 'loc' for my purposes and not
>> the bbox_to_anchor property.
>>
>> Is there really no way to get the dimensions of a legend? It has to be
>> there somewhere, otherwise the legend wouldn't know where to draw
>> itself ;)
>
> Hi Daniel,
>
> I'm replying to the list, so that someone correct me if I'm
> wrong, or point out a better way of doing this.
>
> there totally is a way to get the dimensions of a legend.
>
> You can get it in pixel coordinates using
>
> l = plt.legend()
> bbox = l.get_window_extent()
> bbox.width,bbox.height
>
> or in axes coordinates using something like
>
> bbox2 = bbox.transformed(l.axes.transAxes.inverted())
> bbox2.width,bbox2.height
>
> The bboxes have other handy attributes like p0,p1,x0,x1,y0,y1
> etc, as well as methods like bbox.padded(), etc.
>
> best,
> --
> Paul Ivanov
> 314 address only used for lists, off-list direct email at:
> http://pirsquared.org | GPG/PGP key id: 0x0F3E28F7
>
> ------------------------------------------------------------------------------
> Nokia and AT&T present the 2010 Calling All Innovators-North America contest
> Create new apps & games for the Nokia N8 for consumers in U.S. and Canada
> 10ドル million total in prizes - 4ドルM cash, 500 devices, nearly 6ドルM in marketing
> Develop with Nokia Qt SDK, Web Runtime, or Java and Publish to Ovi Store
> http://p.sf.net/sfu/nokia-dev2dev
> _______________________________________________
> Matplotlib-users mailing list
> Mat...@li...
> https://lists.sourceforge.net/lists/listinfo/matplotlib-users
>
-- 
Daniel Hyams
dh...@gm...

On Thursday, October 28, 2010, Pradyumna <pv...@ps...> wrote:
>
> Hello All,
>
> I am running into same issue except mine won't go away even when I went back to matplotlib 0.99.
>
> Also, something curious I noticed - when I run the examples, I don't get this error message.
>
> Please let me know if you have any suggestions.
>
> Thanks!
> Pradyumna
>
>
> Benjamin Root-2 wrote:
>
> On Sat, Oct 9, 2010 at 7:41 AM, Paul Leopardi wrote:
>
>> On Saturday 09 October 2010 22:58:04 Paul Leopardi wrote:
>> > Hello all
>> > I am seeing a problem similar to that seen by Jorge Scandaliaris.
>>
>> I downgraded from matplotlib 1.0.0 to matplotlib 0.99 and my original
>> problem
>> no longer appears:
>>
>> leopardi@linfinit:~/src/Working/Working-0.5.1/glucat/pyclical> rpm -q -a |
>> grep matplotlib | sort
>> python-matplotlib-0.99.1.1-0.pm.1.8.x86_64
>> leopardi@linfinit:~/src/Working/Working-0.5.1/glucat/pyclical> ipython
>> -pylab
>> Your PyGtk has set_interactive(), so you can use the
>> more stable single-threaded Gtk mode.
>> See https://bugs.launchpad.net/ipython/+bug/270856
>> Python 2.6.2 (r262:71600, Jun 17 2010, 13:37:45)
>> Type "copyright", "credits" or "license" for more information.
>>
>> IPython 0.10 -- An enhanced Interactive Python.
>> ? -> Introduction and overview of IPython's features.
>> %quickref -> Quick reference.
>> help -> Python's own help system.
>> object? -> Details about 'object'. ?object also works, ?? prints more.
>>
>> Welcome to pylab, a matplotlib-based Python environment.
>> For more information, type 'help(pylab)'.
>>
>> In [1]: from mpl_toolkits.mplot3d import Axes3D
>>
>> In [2]: import matplotlib.pyplot as plt
>>
>> In [3]: fig=plt.figure()
>> /usr/lib64/python2.6/site-packages/matplotlib/backends/backend_gtk.py:621:
>> DeprecationWarning: Use the new widget gtk.Tooltip
>> self.tooltips = gtk.Tooltips()
>>
>> In [4]: ax=Axes3D(fig)
>>
>> In [5]: plt.show()
>>
>> In [6]: quit()
>> Do you really want to exit ([y]/n)? y
>> Closing threads... Done.
>>
>> After upgrading to Matplotlib 0.99:
>>
>> leopardi@linfinit:~/src/Working/Working-0.5.1/glucat/pyclical> rpm -q -a |
>> grep matplotlib | sort
>> python-matplotlib-1.0.0-9.2.x86_64
>> python-matplotlib-tk-1.0.0-9.2.x86_64
>> python-matplotlib-wx-1.0.0-9.2.x86_64
>>
>> leopardi@linfinit:~/src/Working/Working-0.5.1/glucat/pyclical> ipython
>> -pylab
>> Python 2.6.2 (r262:71600, Jun 17 2010, 13:37:45)
>> Type "copyright", "credits" or "license" for more information.
>>
>> IPython 0.10 -- An enhanced Interactive Python.
>> ? -> Introduction and overview of IPython's features.
>> %quickref -> Quick reference.
>> help -> Python's own help system.
>> object? -> Details about 'object'. ?object also works, ?? prints more.
>>
>> Welcome to pylab, a matplotlib-based Python environment.
>> For more information, type 'help(pylab)'.
>>
>> In [1]: from mpl_toolkits.mplot3d import Axes3D
>>
>> In [2]: import matplotlib.pyplot as plt
>>
>> In [3]: fig=plt.figure()
>>
>> In [4]: ax=Axes3D(fig)
>>
>> In [5]: plt.show()
>> ^CERROR: An unexpected error occurred while tokenizing input
>> The following traceback may be corrupted or invalid
>> The error message is: ('EOF in multi-line statement', (206, 0))
>>
>> ---------------------------------------------------------------------------
>> KeyboardInterrupt Traceback (most recent call last)
>>
>> /home/leopardi/src/Working/Working-0.5.1/glucat/pyclical/
>> in
>> ()
>>
>> /usr/lib64/python2.6/site-packages/matplotlib/backends/backend_tkagg.pyc in
>> show()
>> 72 for manager in Gcf.get_all_fig_managers():
>> 73 manager.show()
>> ---> 74 Tk.mainloop()
>> 75
>> 76 def new_figure_manager(num, *args, **kwargs):
>>
>> /usr/lib64/python2.6/lib-tk/Tkinter.pyc in mainloop(n)
>> 323 def mainloop(n=0):
>> 324 """Run the main loop of Tcl."""
>> --> 325 _default_root.tk.mainloop(n)
>> 326
>> 327 getint = int
>>
>> KeyboardInterrupt:
>>
>> In [6]:
>> In [6]: quit()
>> Do you really want to exit ([y]/n)? y
>>
>>
> I believe this was a known issue that came about from some fixes made to the
> behavior of show() for the 1.0 release. It was patched shortly thereafter
> and the maintenance branch was also patched. As a workaround, I believe you
> can try one of the other backends or install matplotlib from source.
>
> Does anybody know who maintains the packages for OpenSUSE? It would
> probably be prudent to poke them to update.
>
> Ben Root
>
> ------------------------------------------------------------------------------
> Beautiful is writing same markup. Internet Explorer 9 supports
> standards for HTML5, CSS3, SVG 1.1, ECMAScript5, and DOM L2 & L3.
> Spend less time writing and rewriting code and more time creating great
> experiences on the web. Be a part of the beta today.
> http://p.sf.net/sfu/beautyoftheweb
> _______________________________________________
> Matplotlib-users mailing list
> Mat...@li...
> https://lists.sourceforge.net/lists/listinfo/matplotlib-users
>
>
>
>
>
>
>
> View this message in context: Re: Confirming problem with matplotlib.pyplot.show() <http://old.nabble.com/Confirming-problem-with-matplotlib.pyplot.show%28%29-tp29922229p30080049.html>
> Sent from the matplotlib - users mailing list archive <http://old.nabble.com/matplotlib---users-f2906.html> at Nabble.com.
>
What was the error message? The message in your email was garbled.
Ben Root

2010年10月29日 chewbaccabacca <jss...@gm...>:
> I have three seperate pcolor() plots that I want to merge into one single
> plot (see attached file). Each plot uses a binary color scheme (white and
> blue, white and red, and white and green). [...]
>
> http://old.nabble.com/file/p30087644/contour_phase_boundary.png
You might want to use imshow(),
http://matplotlib.sourceforge.net/api/axes_api.html#matplotlib.axes.Axes.imshow.
Just plot a RGB image. Admittedly you will have to think about how to
manage it to get white instead of black background. Seems this
questions is a problem of how to define the plot task.
Friedrich

2010年10月28日 Benjamin Root <ben...@ou...>:
> On Thu, Oct 28, 2010 at 1:18 PM, Will Grover <wg...@mi...> wrote:
>> Hello matplotlib-users,
>> I'm using subplots to make an array of plots, but because some of the
>> plots have wider y-axis tick labels than others, some of the subplots end up
>> looking too close to each other. Here's an image that shows what I mean:
>>  http://web.mit.edu/wgrover/www/spacing.png
>> I'm currently using pylab.subplots_adjust(hspace = __, vspace = __) to
>> adjust the subplot spacing, but since that applies to all subplots, no one
>> setting looks right for the entire array of plots. Is there any way to set
>> the spacing so that the subplots *plus tick labels* are evenly distributed?
>> Or can I manually specify the spacing between each subplot? Thanks,
>
> It is *possible*, but it is probably would be an incredible amount of work
> (unless someone knows of some nifty trick that I am not aware of). My
> suggestion to make things appear more "even" is to adjust the position of
> the ylabels in the second column to that it takes up more room between the
> first two columns.
> The hard way would be to manually specify the extents in the "position"
> kwarg [left, bottom, width, height] for the constructor of the axes object.
> The values for those four parts would be in the coordinate system of the
> figure object.
And this would not be super-hard. Opposed to the automatical
solution. I'm not familiar with pyplot etc., only with the API, but I
looked it up for you:
http://matplotlib.sourceforge.net/api/pyplot_api.html#matplotlib.pyplot.axes
Notice that the tuple specifies the plot region, i.e. the
"decorations" like labels and tick labels are not included and will be
placed around.
You can also play with the Axes returned by the axes() call. I.e. you
can shift them around, using:
http://matplotlib.sourceforge.net/api/axes_api.html#matplotlib.axes.Axes.set_position.
Friedrich

 On 10/30/10 3:41 AM, John wrote:
> I've searched through the netcdf4-python trunk, but I find no
> reference of NetCDFFile as in how the basemap or Scientific.IO.NetCDF
> modules use it.
>
> For compatibility with my old scripts, is it safe to simply:
>
> from netCDF4 import Dataset as NetCDFFile
>
> ??
>
> Thanks,
> john
>
John: Neither basemap or Scientific.IO.NetCDF uses it. What gave you 
the impression they did? There is a 'NetCDFFile' in function in 
basemap, but it is based on the pure-python netCDF module pupynere. It 
doesn't support netcdf-4 and I only recommend using it for the 
occasional light use.
The netcdf4-python API is mostly compatible with Scientific.IO.NetCDF, 
so if you have old script that use the latter, you can try:
from netCDF4 import Dataset as NetCDFFile
The docs for netcdf4-python are at
http://netcdf4-python.googlecode.com/svn/trunk/docs/netCDF4-module.html
-Jeff

Hi Dharhas,
2010年10月19日 Dharhas Pothina <Dha...@tw...>:
>> There's no way around the ``Decimal``? Otherwise I cannot confirm the
>> inelegancyness except this construct ;-)
> the moneyfmt routine I downloaded requires the Decimal package (i.e decimal.Decimal). I didn't have time to try writing my own version. It has a bug that if you specify number of decimal places = 0 it shows the trailing decimal point.
Is there a specific reason to not use py2.7, as suggested by Goekhan?
I would not bet my life that it's possible without any problems, but
if, then the new format specification seems much better then the
clumsy Decimal "workaround".
I just ran across this while informing myself about py2.7.
Hmm ... I see there is no binary for py2.7 (today), so you'd have to
compile it yourself. Might be kind of a problem if you're using
Windows.
Friedrich

I've searched through the netcdf4-python trunk, but I find no
reference of NetCDFFile as in how the basemap or Scientific.IO.NetCDF
modules use it.
For compatibility with my old scripts, is it safe to simply:
from netCDF4 import Dataset as NetCDFFile
??
Thanks,
john

Daniel Hyams, on 2010年10月29日 23:48, wrote:
> Thanks Paul! Your suggestion got me part of the way, but I've run
> into another problem...I'm using draggable legends, I'm also wanting
> to fetch the current position of the legend after a drag. The
> draggable legend always updates 'loc', and not 'bbox_to_anchor', so
> I'm afraid that I'm stuck manipulating 'loc' for my purposes and not
> the bbox_to_anchor property.
> 
> Is there really no way to get the dimensions of a legend? It has to be
> there somewhere, otherwise the legend wouldn't know where to draw
> itself ;)
Hi Daniel,
I'm replying to the list, so that someone correct me if I'm
wrong, or point out a better way of doing this.
there totally is a way to get the dimensions of a legend.
You can get it in pixel coordinates using 
 l = plt.legend()
 bbox = l.get_window_extent()
 bbox.width,bbox.height
or in axes coordinates using something like
 bbox2 = bbox.transformed(l.axes.transAxes.inverted())
 bbox2.width,bbox2.height
The bboxes have other handy attributes like p0,p1,x0,x1,y0,y1
etc, as well as methods like bbox.padded(), etc.
best, 
-- 
Paul Ivanov
314 address only used for lists, off-list direct email at:
http://pirsquared.org | GPG/PGP key id: 0x0F3E28F7 

Daniel Hyams, on 2010年10月29日 17:07, wrote:
> I realized after sending that off that I need to provide more
> context....sorry about that.
> 
> What I'm trying to do can be boiled down to the following: I'm trying to
> place a legend precisely, using the top left corner of legend as the
> "sticky" point. In other words, if I want to place the legend here:
> 
> +---------------------------------+-----------+
> | | |
> | | legend |
> | The plot... |-----------+
> | |
> | |
> | |
> | |
> | |
> | |
> +---------------------------------+
> 
> I would have thought that I would set bbox_to_anchor = (0,0,1,1), and loc =
> (1,1). I found out quickly, though, that this places the legend like this:
> +-----------+
> | |
> | legend |
> +---------------------------------+-----------+
> | |
> | |
> | The plot... |
> | |
> | |
> | |
> | |
> | |
> | |
> +---------------------------------+
> 
> Which makes perfect sense from matplotlib's perspective. So all I need to
> do is figure out how tall the legend is, and subtract that off the y
> coordinate before passing 'loc' off to matplotlib's legend. I just can't
> seem to figure out how to get that number. I tried
> self.ax.get_legend().get_frame().get_height(), but that just returns 1 all
> the time.
I think you can just get what you want using:
 plt.plot([3,1,4,1,5,9,2,6,5], label='awesome')
 plt.legend(bbox_to_anchor=(1,1),loc=2)
where loc=2 could have also been written as loc='upper left'
> 
> Ascii art is fun! :)
Indeed!
P.S.
Your posts made it to this and the devel list - they
(frustratingly) don't send you your own copy back when you post
something. I usually verify that the post went through my
checking gmane or sourceforge archives.
best,
-- 
Paul Ivanov
314 address only used for lists, off-list direct email at:
http://pirsquared.org | GPG/PGP key id: 0x0F3E28F7 

Alan G Isaac, on 2010年10月28日 21:29, wrote:
> On 10/27/2010 8:21 PM, Paul Ivanov wrote:
> > def onelegend_twinaxes(axis,twin):
> > #make a joint axis legend
> > lines = twin.get_lines()
> > lines.extend(axis.get_lines())
> > labels = [l.get_label() for l in lines]
> > return axis.legend(lines, labels)
> 
> That works.
> 
> > <http://redwood.berkeley.edu/wiki/VS265:_Homework_assignments>
> 
> Cool. What proportion of the students choose Python?
This is the first time that Python's been officially encouraged,
and it's been about 1 in 6.
-- 
Paul Ivanov
314 address only used for lists, off-list direct email at:
http://pirsquared.org | GPG/PGP key id: 0x0F3E28F7 

 On 10/29/10 6:04 PM, John wrote:
> Any good tutorials or examples that people could point me to on
> creating a netcdf file for a grid projected with a stereographic
> projection? It's not clear to me how you would create the lat / lon
> dimensions.
>
> Thanks,
> john
>
>
>
John: The nice thing about netcdf is there are conventions for almost 
any kind of geophysical variable and/or grid.
See http://cf-pcmdi.llnl.gov/documents/cf-conventions/1.4/apf.html
-Jeff

On Fri, Oct 29, 2010 at 7:04 PM, John <was...@gm...> wrote:
> Any good tutorials or examples that people could point me to on
> creating a netcdf file for a grid projected with a stereographic
> projection? It's not clear to me how you would create the lat / lon
> dimensions.
>
> Thanks,
> john
>
>
>
Be careful of saying "irregular" grid when what you really mean is
non-rectilinear. If the data can be represented in a matrix, then chances
are it is regular in some way.
There are a number of schemes for how to organize your data in netcdf. One
way I have seen is to represent the data in a 1D array and then have
parallel arrays of lat and lon coordinates. This works no matter how
irregular the data grid is (particularly nice with sparse data).
If the data can be represented in a regular grid, then you can represent the
data in 2D in the netcdf file, and have parallel arrays of 2-D lat and lon
arrays. The key point of these two methods is that you would have all three
arrays use an index (or two) as the common dimension.
Does that help?
Ben Root

Any good tutorials or examples that people could point me to on
creating a netcdf file for a grid projected with a stereographic
projection? It's not clear to me how you would create the lat / lon
dimensions.
Thanks,
john
-- 
Configuration
``````````````````````````
Basemap: 1.0
Matplotlib: 1.0.0