matplotlib-users Mailing List for matplotlib (Page 12)

Hi Uwe,
Uwe Schmitt wrote:
> Hi,
>
> I'm using
>
> from matplotlib.backends.backend_wx import FigureCanvasWx
>
> for plotting data. It works fine, but the font used on
> Windows XP is not really pretty. I would like to use another
> font family and another size.
I just recently discovered this for a Vista legend problem I had (they 
were to big).
from matplotlib.font_manager import FontProperties
 self.figure.legend((statsLinesP + statsLinesC), (statsTextsP + 
statsTextsC),
 loc=self.legendPos, prop=FontProperties(size='smaller'))
Werner

Hi,
I'm using
 from matplotlib.backends.backend_wx import FigureCanvasWx
for plotting data. It works fine, but the font used on
Windows XP is not really pretty. I would like to use another
font family and another size.
How can I achieve this ?
And another question: I'm using add_subplots mehtod for plotting,
and I would like to increase the vertical space between the
different subplots. Any hint how to do this ?
Greetings, Uwe

A quick fix (since you need an exam, I guess) would be to make the label
with the 'text' command and put it where you want it.
I am interested in an elegant fix,
Mark
On 5/2/07, mat...@li...
>
> Date: Wed, 2 May 2007 11:39:03 -0500
> From: "Ryan Krauss" <rya...@gm...>
> Subject: [Matplotlib-users] creating a blank graph with room for hand
> written tick labels.
> To: matplotlib-users <mat...@li...>
> Message-ID:
> <c5b...@ma...>
> Content-Type: text/plain; charset="iso-8859-1"
>
> I am writing a final exam and I want my students to sketch a graph of
> something and label the plot themselves. So, I need to create an axis
> with x and y labels, but with no tick marks. This I can do, but
> creating blank tick marks seems to get rid of the space where the
> students would write in their own tick marks. Playing with
> xtick.major.pad doesn't seem to help. What should I be doing? This
> is my code:
>
> figure(20)
> clf()
> xlim([0,0.1])
> ylim([-10,10])
> xlabel('Time (sec)')
>
> yticks(arange(-10,11,2))#,['']*10)
>
> ylabel(r'$v_{out}(t)$')
> grid(True)
> fname='blank_time_plot.eps'
> outdir='figs'
>
> ylim([-5,5])
> yticks(arange(-5,5.2),['']*10)
> xticks(arange(1,10),['']*10)
>
> and the result is attached (png).
>
> Thanks,
>
> Ryan
> -------------- next part --------------
> A non-text attachment was scrubbed...
> Name: blank.png
> Type: image/png
> Size: 19059 bytes
> Desc: not available
>
> ------------------------------
>
> -------------------------------------------------------------------------
> This SF.net email is sponsored by DB2 Express
> Download DB2 Express C - the FREE version of DB2 express and take
> control of your XML. No limits. Just data. Click to get it now.
> http://sourceforge.net/powerbar/db2/
>
> ------------------------------
>
> _______________________________________________
> Matplotlib-users mailing list
> Mat...@li...
> https://lists.sourceforge.net/lists/listinfo/matplotlib-users
>
>
> End of Matplotlib-users Digest, Vol 12, Issue 1
> ***********************************************
>

See pylab.figtext()
David D Clark wrote:
> Hello,
>
> I have a figure with four subplots (2x2). I would like to put a title
> centered over the top row. How do I do this?
>
> Thanks,
> Dave
>
> -------------------------------------------------------------------------
> This SF.net email is sponsored by DB2 Express
> Download DB2 Express C - the FREE version of DB2 express and take
> control of your XML. No limits. Just data. Click to get it now.
> http://sourceforge.net/powerbar/db2/
> _______________________________________________
> Matplotlib-users mailing list
> Mat...@li...
> https://lists.sourceforge.net/lists/listinfo/matplotlib-users
>

-----BEGIN PGP SIGNED MESSAGE-----
Hash: SHA1
Hello,
I have a figure with four subplots (2x2). I would like to put a title
centered over the top row. How do I do this?
Thanks,
Dave
-----BEGIN PGP SIGNATURE-----
Version: GnuPG v1.4.5 (Darwin)
Comment: Using GnuPG with Mozilla - http://enigmail.mozdev.org
iD8DBQFGOREuNu7GcwGNZSMRApPnAJ4/I8N5uAHzI+/fFWuMtvmQ8OF9XQCdFXvL
0QF/2m3HgyxkGmhGlPtnzKM=
=MYiT
-----END PGP SIGNATURE-----

"Ryan Krauss" <rya...@gm...> writes:
> I am writing a final exam and I want my students to sketch a graph of
> something and label the plot themselves. So, I need to create an axis
> with x and y labels, but with no tick marks. This I can do, but
> creating blank tick marks seems to get rid of the space where the
> students would write in their own tick marks. 
A quick hack would be to make non-blank tick marks but cause them to
be invisible in another way, e.g. by setting alpha=0 or color='w':
yticks(arange(-5,5.2),['0']*10,alpha=0)
xticks(arange(1,10),['0']*10,alpha=0)
-- 
Jouni K. Seppänen
http://www.iki.fi/jks

I am writing a final exam and I want my students to sketch a graph of
something and label the plot themselves. So, I need to create an axis
with x and y labels, but with no tick marks. This I can do, but
creating blank tick marks seems to get rid of the space where the
students would write in their own tick marks. Playing with
xtick.major.pad doesn't seem to help. What should I be doing? This
is my code:
figure(20)
clf()
xlim([0,0.1])
ylim([-10,10])
xlabel('Time (sec)')
yticks(arange(-10,11,2))#,['']*10)
ylabel(r'$v_{out}(t)$')
grid(True)
fname='blank_time_plot.eps'
outdir='figs'
ylim([-5,5])
yticks(arange(-5,5.2),['']*10)
xticks(arange(1,10),['']*10)
and the result is attached (png).
Thanks,
Ryan

Hello,
My company, Resonon, is releasing our free (as in beer) software 
package, Spectronon. Spectronon is written in python and utilizes 
matplotlib, numpy, PIL, and scipy. Spectronon's main purpose is 
analyzing hyperspectral data (images with many more bands of color than 
the traditional 3: red, green, blue). It can also be useful for more 
traditional image processing. Perhaps of most interest here, it is a 
quick way to get access to a command line with a bunch of useful 
libraries, plotting, and image viewing. It's packaged as a basic exe 
installer. There is also a fairly simple plug-in architecture for 
extending some of Spectronon's behavior. 
 As the main developer, I am very thankful for all the good work that 
gets put in to all these underlying libraries, and I hope that in 
getting this software out for free, it can in turn support more academic 
and research applications of python and its libraries. The release will 
also help me expose bugs, hopefully get more people interested in 
writing plug-ins for Spectronon, and help me learn other ways to improve 
the code base. While the source code is not presently open, the API is 
open in order to support writing plug-ins.
Here are some screenshots:
http://www.spectronon.com/wiki/index.php?title=Screenshots
I recommend users on this list go directly to the developer docs here:
http://www.spectronon.com/wiki/index.php?title=Setup_a_Simple_Development_Environment 
as this tutorial will step you through the aspects of Spectronon of most 
probable interest.
Links to the download, support forums, and more docs are available at 
http://www.spectronon.com
Thanks very much. I hope it's a fun ap for folks, and I greatly 
appreciate any questions, feedback, or comments as we begin to get this 
project off the ground.
Tim

I see. 
 
I only checked that the code in hist() hadn't changed. I guess it pays
to be thorough. Hope this didn't waste too much time.
 
Cheers,
Scott
>>> Eric Firing <ef...@ha...> 4/30/2007 20:02 >>>
Scott,
In svn, barh does accept the 'log' kwarg via **kwargs, and your example
works correctly. Here is the relevant CHANGELOG entry, after 0.90 was
released:
2007年03月03日 Change barh to take a kwargs dict and pass it to bar.
 Fixes sf bug #1669506.
I think it was just a matter of changing the barh signature to:
 def barh(self, bottom, width, height=0.8, left=None, **kwargs):
but I haven't looked back at the svn records to check that.
Eric
Scott Sinclair wrote:
> Hi,
> 
> The following code fails for me with matplotlib-0.90.0 --
> 
> ---------------------------------------------------
> import pylab as pl
> 
> x = pl.randn(1000)
> 
> pl.hist(x, orientation='horizontal')
> 
> pl.show()
> ---------------------------------------------------
> 
> This is because Axes.barh() [called by Axes.hist() in axes.py]
doesn't 
> accept the keyword argument 'log'. The attached patch fixes the
problem 
> for my purposes (linear scaling of the axes) but obviously fails to 
> scale the bin count axis correctly if hist() is called with log=True
> i.e. pl.hist(x, orientation='horizontal', log=True).
> 
> It appears that this is still the case in the latest svn version of
axes.py.
> 
> Cheers,
> Scott
> 
> 
> Please find our Email Disclaimer here-->: _ 
> http://www.ukzn.ac.za/disclaimer
<http://www.ukzn.ac.za/disclaimer/>_
> 
> 
>
------------------------------------------------------------------------
> 
> --- axes.py2007-01-22 18:57:58.000000000 +0200
> +++ patched_axes.py2007-04-30 15:12:15.921875000 +0200
> @@ -4462,7 +4462,7 @@
> if width is None: width = 0.9*(bins[1]-bins[0])
> if orientation == 'horizontal':
> patches = self.barh(bins, n, height=width, left=bottom,
> - align=align, log=log)
> + align=align)
> elif orientation == 'vertical':
> patches = self.bar(bins, n, width=width, bottom=bottom,
> align=align, log=log)
> 
> 
>
------------------------------------------------------------------------
> 
>
-------------------------------------------------------------------------
> This SF.net email is sponsored by DB2 Express
> Download DB2 Express C - the FREE version of DB2 express and take
> control of your XML. No limits. Just data. Click to get it now.
> http://sourceforge.net/powerbar/db2/ 
> 
> 
>
------------------------------------------------------------------------
> 
> _______________________________________________
> Matplotlib-users mailing list
> Mat...@li... 
> https://lists.sourceforge.net/lists/listinfo/matplotlib-users 
Please find our Email Disclaimer here: http://www.ukzn.ac.za/disclaimer/