matplotlib-users Mailing List for matplotlib

On 04/05/2011 10:57 PM, Wolfgang Kerzendorf wrote:
> Dear all,
>
> I have had a quick look at changing the buttons in toolbar2. This seems
> to be backend dependent. It would be really nice to make it possible to
> add buttons to the toolbar and be backend agnostic. The button should
> have a picture and a hovertext associated with it. Is it very hard to
> include this in the next release of matplotlib?
This has been on the wish list for quite a while, but it is not a small 
job. I am not aware that anyone has tackled it.
Eric
>
> Cheers
> Wolfgang
>

On 04/05/2011 11:26 PM, Muffles wrote:
>
> Hello all,
> i have made a python script to plot some netcdf files that go from 4mb to
> 80mb, and the main purpose is to plot one after another. This seems to work
> fine, no problems there, but sometimes the computer crashes and i belive its
> beacause the files are too big. Is there any way to accelerate the plot time
> or reduce memory consumption?
>
> If your asking whats the error, well, actually i get no error, it just never
> finishes plotting. I tried to wait for like 10 hours, and nothing happened
> so i assume it crashed.
>
> Thx in advance
In addition to what Paul wrote, check the following (assuming you are 
using netCDF4 to read the files; the same may apply when using other 
netcdf interfaces):
Use slicing to extract the variables you are plotting as ndarrays rather 
than trying to call mpl routines with the netCDF4 variable objects 
themselves. I haven't tracked down exactly what is happening, but I 
have noticed that extracting and plotting can be orders of magnitude 
faster than feeding the netCDF4 variable directly to the mpl plot command.
Eric

Muffles, on 2011年04月06日 02:26, wrote:
> i have made a python script to plot some netcdf files that go from 4mb to
> 80mb, and the main purpose is to plot one after another. This seems to work
> fine, no problems there, but sometimes the computer crashes and i belive its
> beacause the files are too big. Is there any way to accelerate the plot time
> or reduce memory consumption? 
> 
> If your asking whats the error, well, actually i get no error, it just never
> finishes plotting. I tried to wait for like 10 hours, and nothing happened
> so i assume it crashed.
How many such files would you say you're plotting?
If you are creating multiple figures as you go - are you closing
the ones you no longer need?
If you are using IPython, did you disable the output caching?
I realize this'll be difficult, but if you can reduce the
problematic code to a standalone script that reproduces the
problem, we'd be able to much more definitively track down the
problem and find a solution.
best,
-- 
Paul Ivanov
314 address only used for lists, off-list direct email at:
http://pirsquared.org | GPG/PGP key id: 0x0F3E28F7 

Nico Schlömer, on 2011年04月06日 15:57, wrote:
> given an axes object, is there any way of telling whether
> set_axis_off() has been called on this object?
Hi Nico,
set_axis_off() just sets ax.axison to False, so you could 
look at the value of ax.axison to get the information you need.
best,
-- 
Paul Ivanov
314 address only used for lists, off-list direct email at:
http://pirsquared.org | GPG/PGP key id: 0x0F3E28F7 

Markus Baden, on 2011年04月06日 10:18, wrote:
> Hi,
> 
> I draw four subplots that touch each other. Thus the "middle cross" of the
> frame is drawn twice and appears to be thicker then the "outer rectangle". I
> came across an old post for an custom Axes that would allow to only draw
> part of the frame
> 
> http://www.mail-archive.com/mat...@li.../msg10242.html
> 
> However, when I run the example provided I get an "can't set attribute
> error" (see traceback below). I use Python 2.7.1 |EPD 7.0-1 (32-bit) on Mac
> OS 10.5.8, which includes matplotlib version 1.0.1.
> 
> Is there a fix to get the custom class running again, or is there an
> alternative way to achieve what I intend?
Hi Markus,
I didn't look at that code but I think the alternative you're
looking for is 
ax.spines['right'].set_visible(False)
ax.spines['bottom'].set_visible(False)
where you use the appropriate locations you want changed for the
different subplots axes.
If you're going to have a lot of subplots, you can simplify the
logic by first setting all spines to being invisible, and then
using ax.is_first_col(), is_first_row() and the corresponding
is_last_* methods to set the appropriate spines back to visible.
Like this:
#show only the outside spines
for ax in all_axes:
 for sp in ax.spines.values():
 sp.set_visible(False)
 if ax.is_first_row():
 ax.spines['top'].set_visible(True)
 if ax.is_last_row():
 ax.spines['bottom'].set_visible(True)
 if ax.is_first_col():
 ax.spines['left'].set_visible(True)
 if ax.is_last_col():
 ax.spines['right'].set_visible(True)
best,
-- 
Paul Ivanov
314 address only used for lists, off-list direct email at:
http://pirsquared.org | GPG/PGP key id: 0x0F3E28F7 

Hi all,
given an axes object, is there any way of telling whether
set_axis_off() has been called on this object?
Cheers,
Nico

Hello all,
i have made a python script to plot some netcdf files that go from 4mb to
80mb, and the main purpose is to plot one after another. This seems to work
fine, no problems there, but sometimes the computer crashes and i belive its
beacause the files are too big. Is there any way to accelerate the plot time
or reduce memory consumption? 
If your asking whats the error, well, actually i get no error, it just never
finishes plotting. I tried to wait for like 10 hours, and nothing happened
so i assume it crashed.
Thx in advance
-- 
View this message in context: http://old.nabble.com/Computer-freezes-plotting-tp31331581p31331581.html
Sent from the matplotlib - users mailing list archive at Nabble.com.

Dear all,
I have had a quick look at changing the buttons in toolbar2. This seems 
to be backend dependent. It would be really nice to make it possible to 
add buttons to the toolbar and be backend agnostic. The button should 
have a picture and a hovertext associated with it. Is it very hard to 
include this in the next release of matplotlib?
Cheers
 Wolfgang

Hi,
I draw four subplots that touch each other. Thus the "middle cross" of the
frame is drawn twice and appears to be thicker then the "outer rectangle". I
came across an old post for an custom Axes that would allow to only draw
part of the frame
http://www.mail-archive.com/mat...@li.../msg10242.html
However, when I run the example provided I get an "can't set attribute
error" (see traceback below). I use Python 2.7.1 |EPD 7.0-1 (32-bit) on Mac
OS 10.5.8, which includes matplotlib version 1.0.1.
Is there a fix to get the custom class running again, or is there an
alternative way to achieve what I intend?
Thanks a lot,
Markus
Traceback (most recent call last):
 File "frametest.py", line 165, in <module>
 ax = plt.subplot(sub, projection='frameaxes')
 File
"/Library/Frameworks/Python.framework/Versions/7.0/lib/python2.7/site-packages/matplotlib/pyplot.py",
line 658, in subplot
 a = fig.add_subplot(*args, **kwargs)
 File
"/Library/Frameworks/Python.framework/Versions/7.0/lib/python2.7/site-packages/matplotlib/figure.py",
line 687, in add_subplot
 a = subplot_class_factory(projection_class)(self, *args, **kwargs)
 File
"/Library/Frameworks/Python.framework/Versions/7.0/lib/python2.7/site-packages/matplotlib/axes.py",
line 8380, in __init__
 self._axes_class.__init__(self, fig, self.figbox, **kwargs)
 File
"/Library/Frameworks/Python.framework/Versions/7.0/lib/python2.7/site-packages/matplotlib/axes.py",
line 459, in __init__
 self.cla()
 File "frametest.py", line 138, in cla
 self.frame = self._frame
AttributeError: can't set attribute

On Apr 5, 2011, at 5:51 PM, Jae-Joon Lee wrote:
> On Wed, Apr 6, 2011 at 9:15 AM, Sean Lake <ody...@gm...> wrote:
>> Gary Ruben found the actual bug: math mode doesn't support --.
> 
> Just to clarify, in latex math mode, "$-$" is "-" (minus sign) and
> "$--$" is "--".
> And this is not a bug.
> 
> -JJ
Yes. That is correct.
Andre

On Wed, Apr 6, 2011 at 9:15 AM, Sean Lake <ody...@gm...> wrote:
> Gary Ruben found the actual bug: math mode doesn't support --.
Just to clarify, in latex math mode, "$-$" is "-" (minus sign) and
"$--$" is "--".
And this is not a bug.
-JJ

you appear to have another typo.
> Gary Ruben found the actual bug: math mode doesn't support --. 
"Gary Ruben" --> "Andre Walker-Loud"
:)
On Apr 5, 2011, at 5:15 PM, Sean Lake wrote:
> Ah, sorry about that. In the script I was using it had the closing $. In spite of the typo, Gary Ruben found the actual bug: math mode doesn't support --. 
> 
> Thanks,
> Sean
> 
> On Apr 5, 2011, at 17:08, gary ruben wrote:
> 
>> Um, how about r"80ドル--120$" instead of r"80ドル--120" ?
>> 
>> On Wed, Apr 6, 2011 at 9:56 AM, Sean Lake <ody...@gm...> wrote:
>>> Hello all,
>>> 
>>> I'm trying to specify a range of numbers in a particular legend using LaTeX. In order to do so I'm feeding it the string: r"80ドル--120". The output should be have an endash, "80–120", but I'm getting "80--120". This is a standard feature of LaTeX ( http://en.wikibooks.org/wiki/LaTeX/Formatting ), so I don't know what's going on.
>>> 
>>> Thanks,
>>> Sean Lake
>>> 
>>> uname -a
>>> Darwin dynamic_051.astro.ucla.edu 10.7.0 Darwin Kernel Version 10.7.0: Sat Jan 29 15:17:16 PST 2011; root:xnu-1504937~1/RELEASE_I386 i386
>>> 
>>> (You also have a bug on this web page: http://matplotlib.sourceforge.net/faq/troubleshooting_faq.html#reporting-problems , The line python -c `import matplotlib; print matplotlib.__version__` should not have back-ticks)
>>> /sw/bin/python2.6 -c 'import matplotlib; print matplotlib.__version__'
>>> 1.0.0
>>> 
>>> Got matplotlib via fink:
>>> fink --version
>>> Package manager version: 0.29.21
>>> Distribution version: selfupdate-rsync Sun Apr 3 02:28:24 2011, 10.6, x86_64
>>> Trees: local/main stable/main stable/crypto unstable/main unstable/crypto
>>> 
>>> matplotlibrc file:
>>> text.usetex : True
>>> 
>>> #backend : MacOSX
>>> backend : GTKAgg
>>> #backend : ps
>>> #backend : pdf
>>> 
>>> 
>>> 
>>> ------------------------------------------------------------------------------
>>> Xperia(TM) PLAY
>>> It's a major breakthrough. An authentic gaming
>>> smartphone on the nation's most reliable network.
>>> And it wants your games.
>>> http://p.sf.net/sfu/verizon-sfdev
>>> _______________________________________________
>>> Matplotlib-users mailing list
>>> Mat...@li...
>>> https://lists.sourceforge.net/lists/listinfo/matplotlib-users
>>> 
> 
> 
> ------------------------------------------------------------------------------
> Xperia(TM) PLAY
> It's a major breakthrough. An authentic gaming
> smartphone on the nation's most reliable network.
> And it wants your games.
> http://p.sf.net/sfu/verizon-sfdev
> _______________________________________________
> Matplotlib-users mailing list
> Mat...@li...
> https://lists.sourceforge.net/lists/listinfo/matplotlib-users

Ah, sorry about that. In the script I was using it had the closing $. In spite of the typo, Gary Ruben found the actual bug: math mode doesn't support --. 
Thanks,
Sean
On Apr 5, 2011, at 17:08, gary ruben wrote:
> Um, how about r"80ドル--120$" instead of r"80ドル--120" ?
> 
> On Wed, Apr 6, 2011 at 9:56 AM, Sean Lake <ody...@gm...> wrote:
>> Hello all,
>> 
>> I'm trying to specify a range of numbers in a particular legend using LaTeX. In order to do so I'm feeding it the string: r"80ドル--120". The output should be have an endash, "80–120", but I'm getting "80--120". This is a standard feature of LaTeX ( http://en.wikibooks.org/wiki/LaTeX/Formatting ), so I don't know what's going on.
>> 
>> Thanks,
>> Sean Lake
>> 
>> uname -a
>> Darwin dynamic_051.astro.ucla.edu 10.7.0 Darwin Kernel Version 10.7.0: Sat Jan 29 15:17:16 PST 2011; root:xnu-1504937~1/RELEASE_I386 i386
>> 
>> (You also have a bug on this web page: http://matplotlib.sourceforge.net/faq/troubleshooting_faq.html#reporting-problems , The line python -c `import matplotlib; print matplotlib.__version__` should not have back-ticks)
>> /sw/bin/python2.6 -c 'import matplotlib; print matplotlib.__version__'
>> 1.0.0
>> 
>> Got matplotlib via fink:
>> fink --version
>> Package manager version: 0.29.21
>> Distribution version: selfupdate-rsync Sun Apr 3 02:28:24 2011, 10.6, x86_64
>> Trees: local/main stable/main stable/crypto unstable/main unstable/crypto
>> 
>> matplotlibrc file:
>> text.usetex : True
>> 
>> #backend : MacOSX
>> backend : GTKAgg
>> #backend : ps
>> #backend : pdf
>> 
>> 
>> 
>> ------------------------------------------------------------------------------
>> Xperia(TM) PLAY
>> It's a major breakthrough. An authentic gaming
>> smartphone on the nation's most reliable network.
>> And it wants your games.
>> http://p.sf.net/sfu/verizon-sfdev
>> _______________________________________________
>> Matplotlib-users mailing list
>> Mat...@li...
>> https://lists.sourceforge.net/lists/listinfo/matplotlib-users
>> 

that also doesn't work. The following would work (this is a LaTeX thing - not matplotlib)
r"80--120"
r"80ドル\textrm{--}120$"
andre
On Apr 5, 2011, at 5:08 PM, gary ruben wrote:
> Um, how about r"80ドル--120$" instead of r"80ドル--120" ?
> 
> On Wed, Apr 6, 2011 at 9:56 AM, Sean Lake <ody...@gm...> wrote:
>> Hello all,
>> 
>> I'm trying to specify a range of numbers in a particular legend using LaTeX. In order to do so I'm feeding it the string: r"80ドル--120". The output should be have an endash, "80–120", but I'm getting "80--120". This is a standard feature of LaTeX ( http://en.wikibooks.org/wiki/LaTeX/Formatting ), so I don't know what's going on.
>> 
>> Thanks,
>> Sean Lake
>> 
>> uname -a
>> Darwin dynamic_051.astro.ucla.edu 10.7.0 Darwin Kernel Version 10.7.0: Sat Jan 29 15:17:16 PST 2011; root:xnu-1504937~1/RELEASE_I386 i386
>> 
>> (You also have a bug on this web page: http://matplotlib.sourceforge.net/faq/troubleshooting_faq.html#reporting-problems , The line python -c `import matplotlib; print matplotlib.__version__` should not have back-ticks)
>> /sw/bin/python2.6 -c 'import matplotlib; print matplotlib.__version__'
>> 1.0.0
>> 
>> Got matplotlib via fink:
>> fink --version
>> Package manager version: 0.29.21
>> Distribution version: selfupdate-rsync Sun Apr 3 02:28:24 2011, 10.6, x86_64
>> Trees: local/main stable/main stable/crypto unstable/main unstable/crypto
>> 
>> matplotlibrc file:
>> text.usetex : True
>> 
>> #backend : MacOSX
>> backend : GTKAgg
>> #backend : ps
>> #backend : pdf
>> 
>> 
>> 
>> ------------------------------------------------------------------------------
>> Xperia(TM) PLAY
>> It's a major breakthrough. An authentic gaming
>> smartphone on the nation's most reliable network.
>> And it wants your games.
>> http://p.sf.net/sfu/verizon-sfdev
>> _______________________________________________
>> Matplotlib-users mailing list
>> Mat...@li...
>> https://lists.sourceforge.net/lists/listinfo/matplotlib-users
>> 
> 
> ------------------------------------------------------------------------------
> Xperia(TM) PLAY
> It's a major breakthrough. An authentic gaming
> smartphone on the nation's most reliable network.
> And it wants your games.
> http://p.sf.net/sfu/verizon-sfdev
> _______________________________________________
> Matplotlib-users mailing list
> Mat...@li...
> https://lists.sourceforge.net/lists/listinfo/matplotlib-users

Um, how about r"80ドル--120$" instead of r"80ドル--120" ?
On Wed, Apr 6, 2011 at 9:56 AM, Sean Lake <ody...@gm...> wrote:
> Hello all,
>
> I'm trying to specify a range of numbers in a particular legend using LaTeX. In order to do so I'm feeding it the string: r"80ドル--120". The output should be have an endash, "80–120", but I'm getting "80--120". This is a standard feature of LaTeX ( http://en.wikibooks.org/wiki/LaTeX/Formatting ), so I don't know what's going on.
>
> Thanks,
> Sean Lake
>
> uname -a
> Darwin dynamic_051.astro.ucla.edu 10.7.0 Darwin Kernel Version 10.7.0: Sat Jan 29 15:17:16 PST 2011; root:xnu-1504937~1/RELEASE_I386 i386
>
> (You also have a bug on this web page: http://matplotlib.sourceforge.net/faq/troubleshooting_faq.html#reporting-problems , The line python -c `import matplotlib; print matplotlib.__version__` should not have back-ticks)
> /sw/bin/python2.6 -c 'import matplotlib; print matplotlib.__version__'
> 1.0.0
>
> Got matplotlib via fink:
> fink --version
> Package manager version: 0.29.21
> Distribution version: selfupdate-rsync Sun Apr 3 02:28:24 2011, 10.6, x86_64
> Trees: local/main stable/main stable/crypto unstable/main unstable/crypto
>
> matplotlibrc file:
> text.usetex : True
>
> #backend : MacOSX
> backend : GTKAgg
> #backend : ps
> #backend : pdf
>
>
>
> ------------------------------------------------------------------------------
> Xperia(TM) PLAY
> It's a major breakthrough. An authentic gaming
> smartphone on the nation's most reliable network.
> And it wants your games.
> http://p.sf.net/sfu/verizon-sfdev
> _______________________________________________
> Matplotlib-users mailing list
> Mat...@li...
> https://lists.sourceforge.net/lists/listinfo/matplotlib-users
>

Hi Sean,
I just checked - the hyphenation you want does not work in LaTeX in math mode. Try removing the "$"-signs in your string command. Then the hyphenation should work.
Andre
On Apr 5, 2011, at 4:56 PM, Sean Lake wrote:
> Hello all,
> 
> I'm trying to specify a range of numbers in a particular legend using LaTeX. In order to do so I'm feeding it the string: r"80ドル--120". The output should be have an endash, "80–120", but I'm getting "80--120". This is a standard feature of LaTeX ( http://en.wikibooks.org/wiki/LaTeX/Formatting ), so I don't know what's going on. 
> 
> Thanks,
> Sean Lake
> 
> uname -a
> Darwin dynamic_051.astro.ucla.edu 10.7.0 Darwin Kernel Version 10.7.0: Sat Jan 29 15:17:16 PST 2011; root:xnu-1504937~1/RELEASE_I386 i386
> 
> (You also have a bug on this web page: http://matplotlib.sourceforge.net/faq/troubleshooting_faq.html#reporting-problems , The line python -c `import matplotlib; print matplotlib.__version__` should not have back-ticks)
> /sw/bin/python2.6 -c 'import matplotlib; print matplotlib.__version__'
> 1.0.0
> 
> Got matplotlib via fink:
> fink --version
> Package manager version: 0.29.21
> Distribution version: selfupdate-rsync Sun Apr 3 02:28:24 2011, 10.6, x86_64
> Trees: local/main stable/main stable/crypto unstable/main unstable/crypto
> 
> matplotlibrc file:
> text.usetex : True
> 
> #backend : MacOSX
> backend : GTKAgg
> #backend : ps
> #backend : pdf
> 
> 
> 
> ------------------------------------------------------------------------------
> Xperia(TM) PLAY
> It's a major breakthrough. An authentic gaming
> smartphone on the nation's most reliable network.
> And it wants your games.
> http://p.sf.net/sfu/verizon-sfdev
> _______________________________________________
> Matplotlib-users mailing list
> Mat...@li...
> https://lists.sourceforge.net/lists/listinfo/matplotlib-users