[Matplotlib-users] hatching problem

2010-02-09 Thread Tomasz Koziara

Dear Users/Developers

I just installed version 0.99.1.1 since in my previous version (0.98)  
I had problems with hatching. It seems though that the same problems  
persist in the current version. The attached files reproduce the  
problem (a data file and a python short script). Note that hatching is  
not present on all green 'CONUPD' fields - but only on few of them.


I will appreciate some hints on how to get by,
Regards
Tomek

===


cubes_1_PSEUDO_RIGID_FULL
Description: Binary data



#!/usr/bin/env python
# a stacked bar plot with errorbars
import numpy as np
import matplotlib.pyplot as plt

inp = open (cubes_1_PSEUDO_RIGID_FULL, r)
dat = [[],[],[],[],[],[],[],[],[],[],[]]
lin = inp.readline ()
num = 0
while lin != :
  lst = lin.rsplit ( )
  lin = inp.readline ()
  if len (lst)  2:
dat [num % 11].append (float (lst [2]))
num = num + 1

N = 8
TIMINT = np.array (dat [0])
CONUPD = np.array (dat [1])
CONDET = np.array (dat [2])
LOCDYN = np.array (dat [3])
CONSOL = np.array (dat [4])
PARBAL = np.array (dat [5])
ind = np.arange(N)
width = 0.6

p1 = plt.bar (ind, TIMINT, width, color='r')
p2 = plt.bar (ind, CONUPD, width, color='g', bottom=TIMINT, hatch='//')
p3 = plt.bar (ind, CONDET, width, color='b', bottom=TIMINT+CONUPD, hatch='.')
p4 = plt.bar (ind, LOCDYN, width, color='y', bottom=TIMINT+CONUPD+CONDET, hatch='o')
p5 = plt.bar (ind, CONSOL, width, color='c', bottom=TIMINT+CONUPD+CONDET+LOCDYN, hatch='*')
p6 = plt.bar (ind, PARBAL, width, color='m', bottom=TIMINT+CONUPD+CONDET+LOCDYN+CONSOL, hatch='O')

plt.ylabel ('Time [s]')
plt.title ('Runtimes by processor count')
plt.xticks (ind+width/2., ('1', '2', '4', '8', '16', '32', '64', '128') )
plt.legend ((p6[0], p5[0], p4[0], p3[0], p2[0], p1[0]), ('PARBAL', 'CONSOL', 'LOCDYN', 'CONDET', 'CONUPD', 'TIMEINT'), loc = upper left)
plt.axis (xmin = -0.2, xmax = ind[N-1]+width+0.2)
plt.savefig ('cubes-1-prb-full.eps')
plt.clf ()



===

dh178-192:tkp5 tomek$ python plots.py --verbose-helpful
$HOME=/Users/tomek
CONFIGDIR=/Users/tomek/.matplotlib
matplotlib data path /Library/Python/2.5/site-packages/matplotlib/mpl- 
data
loaded rc file /Library/Python/2.5/site-packages/matplotlib/mpl-data/ 
matplotlibrc

matplotlib version 0.99.1.1
verbose.level helpful
interactive is False
units is False
platform is darwin
Using fontManager instance from /Users/tomek/.matplotlib/fontList.cache
backend MacOSX version unknown
findfont: Matching :family=sans- 
serif:style 
=normal:variant=normal:weight=normal:stretch=normal:size=medium to  
Bitstream Vera Sans (/Library/Python/2.5/site-packages/ 
matplotlib-0.98.6svn-py2.5-macosx-10.3-fat.egg/matplotlib/mpl-data/ 
fonts/ttf/Vera.ttf) with score of 0.00
findfont: Matching :family=sans- 
serif:style 
=normal:variant=normal:weight=normal:stretch=normal:size=large to  
Bitstream Vera Sans (/Library/Python/2.5/site-packages/ 
matplotlib-0.98.6svn-py2.5-macosx-10.3-fat.egg/matplotlib/mpl-data/ 
fonts/ttf/Vera.ttf) with score of 0.00


--
The Planet: dedicated and managed hosting, cloud storage, colocation
Stay online with enterprise data centers and the best network in the business
Choose flexible plans and management services without long-term contracts
Personal 24x7 support from experience hosting pros just a phone call away.
http://p.sf.net/sfu/theplanet-com___
Matplotlib-users mailing list
Matplotlib-users@lists.sourceforge.net
https://lists.sourceforge.net/lists/listinfo/matplotlib-users


Re: [Matplotlib-users] Memory leak when destroying Tk frame containing a figure

2010-02-09 Thread Stephan Markus
I already had my destroy() method look like this:

def destroy(self):
self.f.clf()
Tix.Frame.destroy(self)
self.toolbar.destroy()
self.canvas._tkcanvas.destroy()


But it makes no difference.

Stephan

Am 08.02.2010 17:15, schrieb Michael Droettboom:
 Have you tried explicitly calling .clf() on the matplotlib Figure object from 
 your Tix.Frame.destroy callback?

 Mike

   

--
The Planet: dedicated and managed hosting, cloud storage, colocation
Stay online with enterprise data centers and the best network in the business
Choose flexible plans and management services without long-term contracts
Personal 24x7 support from experience hosting pros just a phone call away.
http://p.sf.net/sfu/theplanet-com
___
Matplotlib-users mailing list
Matplotlib-users@lists.sourceforge.net
https://lists.sourceforge.net/lists/listinfo/matplotlib-users


Re: [Matplotlib-users] hatching problem

2010-02-09 Thread Jae-Joon Lee
I cannot reproduce it with Agg backend and ps backend. I tried both
svn version and 0.99 maint. version.
So, maybe this is a bug in mac os X backend?
Do you see a same problem with the ps output? If so, can you post your
ouput ps file?

Regards,

-JJ


On Mon, Feb 8, 2010 at 3:26 PM, Tomasz Koziara
t.kozi...@civil.gla.ac.uk wrote:
 Dear Users/Developers

 I just installed version 0.99.1.1 since in my previous version (0.98) I had
 problems with hatching. It seems though that the same problems persist in
 the current version. The attached files reproduce the problem (a data file
 and a python short script). Note that hatching is not present on all green
 'CONUPD' fields - but only on few of them.

 I will appreciate some hints on how to get by,
 Regards
 Tomek

 ===






 ===

 dh178-192:tkp5 tomek$ python plots.py --verbose-helpful
 $HOME=/Users/tomek
 CONFIGDIR=/Users/tomek/.matplotlib
 matplotlib data path /Library/Python/2.5/site-packages/matplotlib/mpl-data
 loaded rc file
 /Library/Python/2.5/site-packages/matplotlib/mpl-data/matplotlibrc
 matplotlib version 0.99.1.1
 verbose.level helpful
 interactive is False
 units is False
 platform is darwin
 Using fontManager instance from /Users/tomek/.matplotlib/fontList.cache
 backend MacOSX version unknown
 findfont: Matching
 :family=sans-serif:style=normal:variant=normal:weight=normal:stretch=normal:size=medium
 to Bitstream Vera Sans
 (/Library/Python/2.5/site-packages/matplotlib-0.98.6svn-py2.5-macosx-10.3-fat.egg/matplotlib/mpl-data/fonts/ttf/Vera.ttf)
 with score of 0.00
 findfont: Matching
 :family=sans-serif:style=normal:variant=normal:weight=normal:stretch=normal:size=large
 to Bitstream Vera Sans
 (/Library/Python/2.5/site-packages/matplotlib-0.98.6svn-py2.5-macosx-10.3-fat.egg/matplotlib/mpl-data/fonts/ttf/Vera.ttf)
 with score of 0.00


 --
 The Planet: dedicated and managed hosting, cloud storage, colocation
 Stay online with enterprise data centers and the best network in the
 business
 Choose flexible plans and management services without long-term contracts
 Personal 24x7 support from experience hosting pros just a phone call away.
 http://p.sf.net/sfu/theplanet-com
 ___
 Matplotlib-users mailing list
 Matplotlib-users@lists.sourceforge.net
 https://lists.sourceforge.net/lists/listinfo/matplotlib-users



--
The Planet: dedicated and managed hosting, cloud storage, colocation
Stay online with enterprise data centers and the best network in the business
Choose flexible plans and management services without long-term contracts
Personal 24x7 support from experience hosting pros just a phone call away.
http://p.sf.net/sfu/theplanet-com
___
Matplotlib-users mailing list
Matplotlib-users@lists.sourceforge.net
https://lists.sourceforge.net/lists/listinfo/matplotlib-users


Re: [Matplotlib-users] Figure.draw_artist() bug with Text

2010-02-09 Thread Ben Axelrod
I see.  But why would I need to set the figure manually when I am drawing with 
a figure?  Is it ever the case where you set one figure, but draw with another? 
 For example:

textartist.set_figure(fig1)
fig2.draw_artist(textartist)

Also, other atists don't fail in this manner if I don't use artist.set_figure().

-Ben

-Original Message-
From: Jae-Joon Lee [mailto:lee.j.j...@gmail.com] 
Sent: Monday, February 08, 2010 6:36 PM
To: Ben Axelrod
Cc: matplotlib-users@lists.sourceforge.net
Subject: Re: [Matplotlib-users] Figure.draw_artist() bug with Text

This is not a bug.
The exception is raised simply because textartist.figure is None (and it is 
None because you never set it).
textartist you created is not properly set up (no figure, no axes, no 
transform). You may do

textartist = Text(0.5, 0.5, Foo)
textartist.set_figure(fig)
fig.draw_artist(textartist)
fig.canvas.blit(fig.bbox)

But, this is not the recommended way of doing things.
draw_artist is mainly for doing animation.

http://matplotlib.sourceforge.net/examples/animation/index.html

Regards,

-JJ

On Mon, Feb 8, 2010 at 4:24 PM, Ben Axelrod baxel...@coroware.com wrote:
 I am getting a fault when I try to use Figure.draw_artist() with a 
 matplotlib.text.Text object.  Since matplotlib.text.Text inherits from 
 matplotlib.artist.Artist, which is what draw_artist() takes, this should 
 probably work.

 Tested with latest SVN code on Linux.

 Here is the traceback:

 Traceback (most recent call last):
  File test.py, line 10, in module
    fig.draw_artist(textartist)
  File /usr/local/lib/python2.6/site-packages/matplotlib/figure.py, 
 line 816, in draw_artist
    a.draw(self._cachedRenderer)
  File /usr/local/lib/python2.6/site-packages/matplotlib/artist.py, 
 line 55, in draw_wrapper
    draw(artist, renderer, *kl)
  File /usr/local/lib/python2.6/site-packages/matplotlib/text.py, 
 line 549, in draw
    bbox, info = self._get_layout(renderer)
  File /usr/local/lib/python2.6/site-packages/matplotlib/text.py, 
 line 267, in _get_layout
    key = self.get_prop_tup()
  File /usr/local/lib/python2.6/site-packages/matplotlib/text.py, 
 line 716, in get_prop_tup
    self.figure.dpi, id(self._renderer),
 AttributeError: 'NoneType' object has no attribute 'dpi'

 And here is some simple code to trigger the bug:

 #!/usr/bin/env python
 # display bug in figure.draw_artist(matplotlib.text)
 import matplotlib.pyplot as plt
 from matplotlib.text import Text

 fig = plt.figure()
 plt.draw()

 textartist = Text(0.5, 0.5, Foo)
 fig.draw_artist(textartist)

 plt.show()
 #end code

 Note that I still get the bug even when i specify figsize and dpi on the 
 figure like so:
 fig = plt.figure(figsize=(2,2), dpi=300)

 -Ben
 --
  The Planet: dedicated and managed hosting, cloud storage, 
 colocation Stay online with enterprise data centers and the best 
 network in the business Choose flexible plans and management services 
 without long-term contracts Personal 24x7 support from experience 
 hosting pros just a phone call away.
 http://p.sf.net/sfu/theplanet-com
 ___
 Matplotlib-users mailing list
 Matplotlib-users@lists.sourceforge.net
 https://lists.sourceforge.net/lists/listinfo/matplotlib-users


--
The Planet: dedicated and managed hosting, cloud storage, colocation
Stay online with enterprise data centers and the best network in the business
Choose flexible plans and management services without long-term contracts
Personal 24x7 support from experience hosting pros just a phone call away.
http://p.sf.net/sfu/theplanet-com
___
Matplotlib-users mailing list
Matplotlib-users@lists.sourceforge.net
https://lists.sourceforge.net/lists/listinfo/matplotlib-users


Re: [Matplotlib-users] Verifying the Use of show()? Win XP

2010-02-09 Thread Wayne Watson
Version of Python would help too. I'm using 2.5.

On 2/9/2010 8:06 AM, Wayne Watson wrote:
 I'm sure not making much progress on understanding show(). When used in
 XP in IDLE or by file execution (click on file name), it seems to tie up
 the executing program.  In IDLE, the shell window stops and one must
 exit the window.

 I'd appreciate it if someone could take any examples from
 http://matplotlib.sourceforge.net/index.html  and try to execute them
 as in the first paragraph to see if they terminate successfully.Let me
 know what OS used, hopefully XP, and if you used IDLE or file execution.
 I suspect you will find every example there ends with show(). Try
 putting a print statement after show() you've done it with the show()
 the last line.


-- 
Crime is way down. War is declining. And that's far from the good 
news. -- Steven Pinker (and other sources) Why is this true, but yet 
the media says otherwise? The media knows very well how to manipulate us 
(see limbic, emotion, $$). -- WTW

--
The Planet: dedicated and managed hosting, cloud storage, colocation
Stay online with enterprise data centers and the best network in the business
Choose flexible plans and management services without long-term contracts
Personal 24x7 support from experience hosting pros just a phone call away.
http://p.sf.net/sfu/theplanet-com
___
Matplotlib-users mailing list
Matplotlib-users@lists.sourceforge.net
https://lists.sourceforge.net/lists/listinfo/matplotlib-users


Re: [Matplotlib-users] Verifying the Use of show()? Win XP

2010-02-09 Thread John Hunter
On Tue, Feb 9, 2010 at 10:06 AM, Wayne Watson
sierra_mtnv...@sbcglobal.net wrote:
 I'm sure not making much progress on understanding show(). When used in
 XP in IDLE or by file execution (click on file name), it seems to tie up
 the executing program.  In IDLE, the shell window stops and one must
 exit the window.

 I'd appreciate it if someone could take any examples from
 http://matplotlib.sourceforge.net/index.html and try to execute them
 as in the first paragraph to see if they terminate successfully.Let me
 know what OS used, hopefully XP, and if you used IDLE or file execution.
 I suspect you will find every example there ends with show(). Try
 putting a print statement after show() you've done it with the show()
 the last line.

show is meant to start the GUI mainloop, which is usually blocking,
and raise all windows, so the behavior you are reporting is the
intended behavior.  When working interactively, as in Idle, you
shouldn't need to use show if you turn interactive mode on.

  http://matplotlib.sourceforge.net/users/shell.html

We recommend using ipython in pylab model when working interactively
because it is designed to make the correct interactive settings and
override show to be non-blocking.  You can obtain the right results
in matplotlib using Idle if you are careful, but for just works out
of the box ipython in pylab mode will be easier.

JDH

--
The Planet: dedicated and managed hosting, cloud storage, colocation
Stay online with enterprise data centers and the best network in the business
Choose flexible plans and management services without long-term contracts
Personal 24x7 support from experience hosting pros just a phone call away.
http://p.sf.net/sfu/theplanet-com
___
Matplotlib-users mailing list
Matplotlib-users@lists.sourceforge.net
https://lists.sourceforge.net/lists/listinfo/matplotlib-users


Re: [Matplotlib-users] Verifying the Use of show()? Win XP

2010-02-09 Thread Ben Axelrod
Maybe instead of plot.show() you should do something like:

plot.draw()
raw_input('Press ENTER to exit')

Personally, I also use IDLE on Windows XP to edit my matplotlib files.  
However, I never execute in IDLE.  I simply double click the file in windows 
explorer.

-Ben

-Original Message-
From: Wayne Watson [mailto:sierra_mtnv...@sbcglobal.net] 
Sent: Tuesday, February 09, 2010 11:07 AM
To: matplotlib-users@lists.sourceforge.net
Subject: [Matplotlib-users] Verifying the Use of show()? Win XP

I'm sure not making much progress on understanding show(). When used in XP in 
IDLE or by file execution (click on file name), it seems to tie up the 
executing program.  In IDLE, the shell window stops and one must exit the 
window.

I'd appreciate it if someone could take any examples from 
http://matplotlib.sourceforge.net/index.html and try to execute them as in 
the first paragraph to see if they terminate successfully.Let me know what OS 
used, hopefully XP, and if you used IDLE or file execution. 
I suspect you will find every example there ends with show(). Try putting a 
print statement after show() you've done it with the show() the last line.
--
Crime is way down. War is declining. And that's far from the good news. -- 
Steven Pinker (and other sources) Why is this true, but yet the media says 
otherwise? The media knows very well how to manipulate us (see limbic, emotion, 
$$). -- WTW

--
The Planet: dedicated and managed hosting, cloud storage, colocation Stay 
online with enterprise data centers and the best network in the business Choose 
flexible plans and management services without long-term contracts Personal 
24x7 support from experience hosting pros just a phone call away.
http://p.sf.net/sfu/theplanet-com
___
Matplotlib-users mailing list
Matplotlib-users@lists.sourceforge.net
https://lists.sourceforge.net/lists/listinfo/matplotlib-users

--
The Planet: dedicated and managed hosting, cloud storage, colocation
Stay online with enterprise data centers and the best network in the business
Choose flexible plans and management services without long-term contracts
Personal 24x7 support from experience hosting pros just a phone call away.
http://p.sf.net/sfu/theplanet-com
___
Matplotlib-users mailing list
Matplotlib-users@lists.sourceforge.net
https://lists.sourceforge.net/lists/listinfo/matplotlib-users


[Matplotlib-users] his problems...

2010-02-09 Thread Nick Schurch
HI all,

I've been using matplotlip for a while now but mainly for line plots,
scatter plots and the odd dendrogram. I recently tried plotting a
histogram (of a binomial function) and encountered a problem. So I
though I'd try the extremely simple example set on the front of the
matplotlib page and heres what I got:

Python 2.4.3 (#1, Sep  3 2009, 15:37:12)
[GCC 4.1.2 20080704 (Red Hat 4.1.2-46)] on linux2
Type help, copyright, credits or license for more information.
 from pylab import randn, hist
 x = randn(1)
 hist(x, 100)
Traceback (most recent call last):
  File stdin, line 1, in ?
  File /usr/lib/python2.4/site-packages/matplotlib/pyplot.py, line
1633, in hist
ret =  gca().hist(*args, **kwargs)
  File /usr/lib/python2.4/site-packages/matplotlib/axes.py, line 5060, in hist
align=align, log=log)
  File /usr/lib/python2.4/site-packages/matplotlib/axes.py, line 3253, in bar
assert len(height)==nbars, argument 'height' must be %d or scalar % nbars
AssertionError: argument 'height' must be 101 or scalar

Any idea why this isn't working? I have matplotlib v0.91.2 - will
updating to 0.99 solve the problem?
-- 
Cheers,

Nick Schurch

--
The Planet: dedicated and managed hosting, cloud storage, colocation
Stay online with enterprise data centers and the best network in the business
Choose flexible plans and management services without long-term contracts
Personal 24x7 support from experience hosting pros just a phone call away.
http://p.sf.net/sfu/theplanet-com
___
Matplotlib-users mailing list
Matplotlib-users@lists.sourceforge.net
https://lists.sourceforge.net/lists/listinfo/matplotlib-users


Re: [Matplotlib-users] Figure.draw_artist() bug with Text

2010-02-09 Thread Jae-Joon Lee
Figure.draw_artist is just a convenience function.

def draw_artist(self, a):

draw :class:`matplotlib.artist.Artist` instance *a* only --
this is available only after the figure is drawn

assert self._cachedRenderer is not None
a.draw(self._cachedRenderer)

And, all the drawing is done by the artist itself.



 Also, other atists don't fail in this manner if I don't use 
 artist.set_figure().


As far as I can see, it only means that you're lucky with other artists.
The reason Text artists needs a reference to the figure is to access
the dpi value, as the exception indicates.
All artists are meant to be added to an axes (or at least axes,
figure, etc. attributes set appropriately) to work correctly.

Regards,

-JJ

--
The Planet: dedicated and managed hosting, cloud storage, colocation
Stay online with enterprise data centers and the best network in the business
Choose flexible plans and management services without long-term contracts
Personal 24x7 support from experience hosting pros just a phone call away.
http://p.sf.net/sfu/theplanet-com
___
Matplotlib-users mailing list
Matplotlib-users@lists.sourceforge.net
https://lists.sourceforge.net/lists/listinfo/matplotlib-users


Re: [Matplotlib-users] trying to Copy plot from one MplWidget canvas to another

2010-02-09 Thread Jae-Joon Lee
Copying a matplotlib canvas (or a figure, or an axes) is not easy. You
cannot just rebind it. You need to copy all the hierarchy of
underlying artists. Also the attributes of artists need to be adjusted
accordingly.
And best option in my opinion is just to create another canvas using
the code that created the original canvas.

Regards,

-JJ



On Mon, Feb 8, 2010 at 3:04 PM,  dave.m.t...@seagate.com wrote:
 Hey folks,

 my problem may be obvious, but i can't seem to copy a plot from one canvas
 to another.

 # I have this object where whichCanvas is an instance of MplWidget (code
 shown below)

 self.whichCanvas.canvas.ax.plot(xData, yData, 'bo', linewidth=1.5,
 linestyle='-')




 # I want to copy the plot and axes to another MplWidget object

 self.anotherCanvas.canvas



 I've tried:

 self.anotherCanvas.canvas.ax = self.whichCanvas.canvas.ax
 self.anotherCanvas.canvas.draw()

 and

 self.anotherCanvas.canvas = self.whichCanvas.canvas
 self.anotherCanvas.canvas.draw()

 the plot doesn't seem to copy. Does any body more familiar with matplotlib
 have any suggestions?





 This is the MplWidget Class
 * MplWidget ***

 from PyQt4 import QtGui
 from matplotlib.backends.backend_qt4agg \
 import FigureCanvasQTAgg as FigureCanvas
 from matplotlib.figure import Figure




 class MplCanvas(FigureCanvas):
 def __init__(self):
 self.fig = Figure()
 self.ax = self.fig.add_subplot(111)
 FigureCanvas.__init__(self, self.fig)
 FigureCanvas.setSizePolicy(self,
 QtGui.QSizePolicy.Expanding,
 QtGui.QSizePolicy.Expanding)
 FigureCanvas.updateGeometry(self)




 class MplWidget(QtGui.QWidget):

 def __init__(self, parent = None):
 QtGui.QWidget.__init__(self, parent)
 self.canvas = MplCanvas()
 self.vbl = QtGui.QVBoxLayout()
 self.vbl.addWidget(self.canvas)
 self.setLayout(self.vbl)



 Regards,
 Dave Tung

 cell: 925-321-6657
 office: 510-353-4770
 dave.m.t...@seagate.com

 --
 The Planet: dedicated and managed hosting, cloud storage, colocation
 Stay online with enterprise data centers and the best network in the
 business
 Choose flexible plans and management services without long-term contracts
 Personal 24x7 support from experience hosting pros just a phone call away.
 http://p.sf.net/sfu/theplanet-com
 ___
 Matplotlib-users mailing list
 Matplotlib-users@lists.sourceforge.net
 https://lists.sourceforge.net/lists/listinfo/matplotlib-users



--
The Planet: dedicated and managed hosting, cloud storage, colocation
Stay online with enterprise data centers and the best network in the business
Choose flexible plans and management services without long-term contracts
Personal 24x7 support from experience hosting pros just a phone call away.
http://p.sf.net/sfu/theplanet-com
___
Matplotlib-users mailing list
Matplotlib-users@lists.sourceforge.net
https://lists.sourceforge.net/lists/listinfo/matplotlib-users


Re: [Matplotlib-users] Verifying the Use of show()? Win XP

2010-02-09 Thread Wayne Watson
Thanks for the clarification. I'm really new to this, and do not know of 
interactive mode. I presume there's a command to turn it off and on, and 
possibly an example on how to do it. Is this method of operation with 
show() mentioned anywhere?

The unfortunate present use of show() is that it ties up the shell 
script, where I happened to have written program output. It's handy to 
put it there, since it's meant to be interactive.  The user is keyboard 
arrowing through images, and statistical data is placed on the shell 
window. At the same time he sees a plot of data relevant to the image. 
He needs to close the plot window before going to the next image. I can 
probably figure out how to kill the plot window when he does that.

My problem with using ipython is that the program I'm modifying is used 
with IDLE, and people have gotten to use it that way. I had nothing to 
do with that method of op. I doubt any of the users would be agreeable 
to using ipython. None of them know Python. The next time the program is 
released, I may provide it in executable form.

I used matlab five years ago, for about two months. To see if it could 
help me understand MPL, I fired it up, and it's now working. Perhaps the 
interactive op is explained there. I take it there is no show() there?

Interesting mention of non-blocking. In the midst of this dilemma, I 
started getting socket errors. Using McAffe I found pythonw as blocked. 
Would that be in anyway associated with the use of show()?   I've since 
changed it to outbound blocking.

On 2/9/2010 8:18 AM, John Hunter wrote:
 On Tue, Feb 9, 2010 at 10:06 AM, Wayne Watson
 sierra_mtnv...@sbcglobal.net  wrote:

...
 the last line.

 show is meant to start the GUI mainloop, which is usually blocking,
 and raise all windows, so the behavior you are reporting is the
 intended behavior.  When working interactively, as in Idle, you
 shouldn't need to use show if you turn interactive mode on.

http://matplotlib.sourceforge.net/users/shell.html

 We recommend using ipython in pylab model when working interactively
 because it is designed to make the correct interactive settings and
 override show to be non-blocking.  You can obtain the right results
 in matplotlib using Idle if you are careful, but for just works out
 of the box ipython in pylab mode will be easier.

 JDH



-- 
Crime is way down. War is declining. And that's far from the good 
news. -- Steven Pinker (and other sources) Why is this true, but yet 
the media says otherwise? The media knows very well how to manipulate us 
(see limbic, emotion, $$). -- WTW

--
The Planet: dedicated and managed hosting, cloud storage, colocation
Stay online with enterprise data centers and the best network in the business
Choose flexible plans and management services without long-term contracts
Personal 24x7 support from experience hosting pros just a phone call away.
http://p.sf.net/sfu/theplanet-com
___
Matplotlib-users mailing list
Matplotlib-users@lists.sourceforge.net
https://lists.sourceforge.net/lists/listinfo/matplotlib-users


Re: [Matplotlib-users] Verifying the Use of show()? Win XP

2010-02-09 Thread Wayne Watson
Thanks. I'll consider it. See my post to John Hunter moments ago as to 
why I must use IDLE. Basically, the users are not at all familiar with 
Python, and historically the us has been that way for 3-4 years.

On 2/9/2010 8:20 AM, Ben Axelrod wrote:
 Maybe instead of plot.show() you should do something like:

 plot.draw()
 raw_input('Press ENTER to exit')

 Personally, I also use IDLE on Windows XP to edit my matplotlib files.  
 However, I never execute in IDLE.  I simply double click the file in windows 
 explorer.

 -Ben

 -Original Message-
 From: Wayne Watson [mailto:sierra_mtnv...@sbcglobal.net]
 Sent: Tuesday, February 09, 2010 11:07 AM
 To: matplotlib-users@lists.sourceforge.net
 Subject: [Matplotlib-users] Verifying the Use of show()? Win XP

 I'm sure not making much progress on understanding show(). When used in XP in 
 IDLE or by file execution (click on file name), it seems to tie up the 
 executing program.  In IDLE, the shell window stops and one must exit the 
 window.

 I'd appreciate it if someone could take any examples 
 fromhttp://matplotlib.sourceforge.net/index.html  and try to execute them 
 as in the first paragraph to see if they terminate successfully.Let me know 
 what OS used, hopefully XP, and if you used IDLE or file execution.
 I suspect you will find every example there ends with show(). Try putting a 
 print statement after show() you've done it with the show() the last line.
 --
 Crime is way down. War is declining. And that's far from the good news. -- 
 Steven Pinker (and other sources) Why is this true, but yet the media says 
 otherwise? The media knows very well how to manipulate us (see limbic, 
 emotion, $$). -- WTW

 --
 The Planet: dedicated and managed hosting, cloud storage, colocation Stay 
 online with enterprise data centers and the best network in the business 
 Choose flexible plans and management services without long-term contracts 
 Personal 24x7 support from experience hosting pros just a phone call away.
 http://p.sf.net/sfu/theplanet-com
 ___
 Matplotlib-users mailing list
 Matplotlib-users@lists.sourceforge.net
 https://lists.sourceforge.net/lists/listinfo/matplotlib-users



-- 
Crime is way down. War is declining. And that's far from the good 
news. -- Steven Pinker (and other sources) Why is this true, but yet 
the media says otherwise? The media knows very well how to manipulate us 
(see limbic, emotion, $$). -- WTW

--
The Planet: dedicated and managed hosting, cloud storage, colocation
Stay online with enterprise data centers and the best network in the business
Choose flexible plans and management services without long-term contracts
Personal 24x7 support from experience hosting pros just a phone call away.
http://p.sf.net/sfu/theplanet-com
___
Matplotlib-users mailing list
Matplotlib-users@lists.sourceforge.net
https://lists.sourceforge.net/lists/listinfo/matplotlib-users


Re: [Matplotlib-users] Verifying the Use of show()? Win XP

2010-02-09 Thread John Hunter
On Tue, Feb 9, 2010 at 11:36 AM, Wayne Watson
sierra_mtnv...@sbcglobal.net wrote:
 Thanks for the clarification. I'm really new to this, and do not know of
 interactive mode. I presume there's a command to turn it off and on, and
 possibly an example on how to do it. Is this method of operation with
 show() mentioned anywhere?

Yes, on the documentation link I referred you to.

JDH

--
The Planet: dedicated and managed hosting, cloud storage, colocation
Stay online with enterprise data centers and the best network in the business
Choose flexible plans and management services without long-term contracts
Personal 24x7 support from experience hosting pros just a phone call away.
http://p.sf.net/sfu/theplanet-com
___
Matplotlib-users mailing list
Matplotlib-users@lists.sourceforge.net
https://lists.sourceforge.net/lists/listinfo/matplotlib-users


Re: [Matplotlib-users] Verifying the Use of show()? Win XP

2010-02-09 Thread Wayne Watson
Thanks. Completely missed it. (P.S. Reply All +corrected mail filter.)

On 2/9/2010 8:18 AM, John Hunter wrote:
 On Tue, Feb 9, 2010 at 10:06 AM, Wayne Watson
 sierra_mtnv...@sbcglobal.net  wrote:

 I'm sure not making much progress on understanding show(). When used in
 XP in IDLE or by file execution (click on file name), it seems to tie up
 the executing program.  In IDLE, the shell window stops and one must
  
...
http://matplotlib.sourceforge.net/users/shell.html

 We recommend using ipython in pylab model when working interactively
 because it is designed to make the correct interactive settings and
 override show to be non-blocking.  You can obtain the right results
 in matplotlib using Idle if you are careful, but for just works out
 of the box ipython in pylab mode will be easier.

 JDH



-- 
Crime is way down. War is declining. And that's far from the good 
news. -- Steven Pinker (and other sources) Why is this true, but yet 
the media says otherwise? The media knows very well how to manipulate us 
(see limbic, emotion, $$). -- WTW

--
The Planet: dedicated and managed hosting, cloud storage, colocation
Stay online with enterprise data centers and the best network in the business
Choose flexible plans and management services without long-term contracts
Personal 24x7 support from experience hosting pros just a phone call away.
http://p.sf.net/sfu/theplanet-com
___
Matplotlib-users mailing list
Matplotlib-users@lists.sourceforge.net
https://lists.sourceforge.net/lists/listinfo/matplotlib-users


Re: [Matplotlib-users] Runtime Error - Need Advice

2010-02-09 Thread Christoph Gohlke
Change the last line to pylab.show() and it should work.

Anyway, this example should not crash the interpreter. I can reproduce
the crash on Python 2.4, 2.5, and 2.6 (32 and 64 bit) on Windows with
mpl 0.99.1 but not on Ubuntu 9.1 with mpl 0.99.0.

The shortest example that crashes is:

python -c import pylab;pylab.subplot(111).figure.show()

or on the interactive prompt:

 import pylab
 pylab.subplot(111).figure.show()
 exit()
Fatal Python error: PyEval_RestoreThread: NULL tstate

This application has requested the Runtime to terminate it in an unusual
way.
Please contact the application's support team for more information.


-- Christoph

On 2/9/2010 7:57 AM, Lee Boger wrote:
  
  Windows XP Professional with Python 2.5 installed (pywin32 build 210) -
  came with dSPACE software package
  
  Downloaded and installed matplotlib-0.99.1.win32-py2.5.exe from
  sourceforge.net
  
  Downloaded and installed numpy-1.4.0-win32-superpack-python2.5.exe from
  sourceforge.net
  
  Executing the following simple log plot script within PythonWin:
  
  *from* matplotlib *import* pylab
  
  # Create some artificial data.
  test_frequency = [1, 2, 3, 4, 5, 6, 7, 8, 9, 10, 11, 12, 13, 14, 15, 16,
  17, 18, 19, 20]
  test_results = [-0.2, -0.7, -1.0, -1.5, -2.0, -2.5, -3.0, -3.5, -4, -5,
  -6, -7.1, -8, -9, -10, -11, -12, -15, -20, -25]
  spec_frequency = [6, 8]
  spec_results = [-3.0, -3.0]
  
  # Plot
  figure = pylab.subplot(111)
  figure.semilogx()
  figure.scatter(test_frequency, test_results, s=20, c='b', marker='s',
  edgecolors='none')
  figure.scatter(spec_frequency, spec_results, s=40, c='g', marker='s',
  edgecolors='none')
  figure.grid(True)
  figure.set_xlabel(rFrequency (Hz), fontsize = 12)
  figure.set_ylabel(rActuator Response (db), fontsize = 12)
  
  figure.figure.savefig('log_plot')
  figure.figure.show()
  
  
  Plots a figure on the screen that looks correct, then the following
  error (when I click OK, PythonWin closes)
  
  
  
  Any advice would be appreciated. Maybe it's an installation or setup
  issue, but I'm pretty knew to Python programming and don't know how to
  debug this.
  
  Lee Boger


--
The Planet: dedicated and managed hosting, cloud storage, colocation
Stay online with enterprise data centers and the best network in the business
Choose flexible plans and management services without long-term contracts
Personal 24x7 support from experience hosting pros just a phone call away.
http://p.sf.net/sfu/theplanet-com
___
Matplotlib-users mailing list
Matplotlib-users@lists.sourceforge.net
https://lists.sourceforge.net/lists/listinfo/matplotlib-users


Re: [Matplotlib-users] Runtime Error - Need Advice

2010-02-09 Thread Eric Firing
Lee Boger wrote:
 
 Windows XP Professional with Python 2.5 installed (pywin32 build 210) - 
 came with dSPACE software package
 
 Downloaded and installed matplotlib-0.99.1.win32-py2.5.exe from 
 sourceforge.net
 
 Downloaded and installed numpy-1.4.0-win32-superpack-python2.5.exe from 
 sourceforge.net

I suspect this is the problem: numpy-1.4.0 is considered broken and has 
been withdrawn.  Specifically, it introduces binary incompatibility with 
programs compiled against prior versions of numpy, including matplotlib.
If you go back to the numpy sourceforge site now, I think you will see 
an earlier version that you can install in place of 1.4.0 (after 
removing the latter).

Eric

 
 Executing the following simple log plot script within PythonWin:
 
 *from* matplotlib *import* pylab
 
 # Create some artificial data.
 test_frequency = [1, 2, 3, 4, 5, 6, 7, 8, 9, 10, 11, 12, 13, 14, 15, 16, 
 17, 18, 19, 20]
 test_results = [-0.2, -0.7, -1.0, -1.5, -2.0, -2.5, -3.0, -3.5, -4, -5, 
 -6, -7.1, -8, -9, -10, -11, -12, -15, -20, -25]
 spec_frequency = [6, 8]
 spec_results = [-3.0, -3.0]
 
 # Plot
 figure = pylab.subplot(111)
 figure.semilogx()
 figure.scatter(test_frequency, test_results, s=20, c='b', marker='s', 
 edgecolors='none')
 figure.scatter(spec_frequency, spec_results, s=40, c='g', marker='s', 
 edgecolors='none')
 figure.grid(True)
 figure.set_xlabel(rFrequency (Hz), fontsize = 12)
 figure.set_ylabel(rActuator Response (db), fontsize = 12)
 
 figure.figure.savefig('log_plot')
 figure.figure.show()
 
 
 Plots a figure on the screen that looks correct, then the following 
 error (when I click OK, PythonWin closes)
 
 
 
 Any advice would be appreciated. Maybe it's an installation or setup 
 issue, but I'm pretty knew to Python programming and don't know how to 
 debug this.
 
 Lee Boger

--
The Planet: dedicated and managed hosting, cloud storage, colocation
Stay online with enterprise data centers and the best network in the business
Choose flexible plans and management services without long-term contracts
Personal 24x7 support from experience hosting pros just a phone call away.
http://p.sf.net/sfu/theplanet-com
___
Matplotlib-users mailing list
Matplotlib-users@lists.sourceforge.net
https://lists.sourceforge.net/lists/listinfo/matplotlib-users


Re: [Matplotlib-users] Runtime Error - Need Advice

2010-02-09 Thread Christoph Gohlke
Use pylab.savefig('log_plot'). Also saving the figure to file and
showing it interactively in the same script might involve switching
backends, which might not work as expected.

It is advisable to downgrade to numpy 1.3.

-- Christoph

On 2/9/2010 10:44 AM, Lee Boger wrote:
 
 Although, if I close the figure then re-run the script, a new figure
 pops up but it doesn't have any data plotted. Interpreter is now locked
 up. I'm still not fixing it completely. There is also no figure stored
 as a file log_plot.
 
 Lee
 
 
 *Christoph Gohlke cgoh...@uci.edu*
 
 02/09/2010 01:23 PM
 
   
 To
   Lee Boger boger_...@cat.com
 cc
   
 Subject
   Re: [Matplotlib-users] Runtime Error - Need Advice
 
 
   
 
 
 Caterpillar: Confidential GreenRetain Until: 03/11/2010
 
 
 
 
 Change the last line to pylab.show() and it should work.
 
 Anyway, this example should not crash the interpreter. I can reproduce
 the crash on Python 2.4, 2.5, and 2.6 (32 and 64 bit) on Windows with
 mpl 0.99.1 but not on Ubuntu 9.1 with mpl 0.99.0.
 
 The shortest example that crashes is:
 
 python -c import pylab;pylab.subplot(111).figure.show()
 
 or on the interactive prompt:
 
 import pylab
 pylab.subplot(111).figure.show()
 exit()
 Fatal Python error: PyEval_RestoreThread: NULL tstate
 
 This application has requested the Runtime to terminate it in an unusual
 way.
 Please contact the application's support team for more information.
 
 
 -- Christoph
 
 On 2/9/2010 7:57 AM, Lee Boger wrote:

 Windows XP Professional with Python 2.5 installed (pywin32 build 210) -
 came with dSPACE software package

 Downloaded and installed matplotlib-0.99.1.win32-py2.5.exe from
 sourceforge.net

 Downloaded and installed numpy-1.4.0-win32-superpack-python2.5.exe from
 sourceforge.net

 Executing the following simple log plot script within PythonWin:

 *from* matplotlib *import* pylab

 # Create some artificial data.
 test_frequency = [1, 2, 3, 4, 5, 6, 7, 8, 9, 10, 11, 12, 13, 14, 15, 16,
 17, 18, 19, 20]
 test_results = [-0.2, -0.7, -1.0, -1.5, -2.0, -2.5, -3.0, -3.5, -4, -5,
 -6, -7.1, -8, -9, -10, -11, -12, -15, -20, -25]
 spec_frequency = [6, 8]
 spec_results = [-3.0, -3.0]

 # Plot
 figure = pylab.subplot(111)
 figure.semilogx()
 figure.scatter(test_frequency, test_results, s=20, c='b', marker='s',
 edgecolors='none')
 figure.scatter(spec_frequency, spec_results, s=40, c='g', marker='s',
 edgecolors='none')
 figure.grid(True)
 figure.set_xlabel(rFrequency (Hz), fontsize = 12)
 figure.set_ylabel(rActuator Response (db), fontsize = 12)

 figure.figure.savefig('log_plot')
 figure.figure.show()


 Plots a figure on the screen that looks correct, then the following
 error (when I click OK, PythonWin closes)



 Any advice would be appreciated. Maybe it's an installation or setup
 issue, but I'm pretty knew to Python programming and don't know how to
 debug this.

 Lee Boger



 --
 The Planet: dedicated and managed hosting, cloud storage, colocation
 Stay online with enterprise data centers and the best network in the
 business
 Choose flexible plans and management services without long-term contracts
 Personal 24x7 support from experience hosting pros just a phone call away.
 http://p.sf.net/sfu/theplanet-com



 ___
 Matplotlib-users mailing list
 Matplotlib-users@lists.sourceforge.net
 https://lists.sourceforge.net/lists/listinfo/matplotlib-users
 

--
The Planet: dedicated and managed hosting, cloud storage, colocation
Stay online with enterprise data centers and the best network in the business
Choose flexible plans and management services without long-term contracts
Personal 24x7 support from experience hosting pros just a phone call away.
http://p.sf.net/sfu/theplanet-com
___
Matplotlib-users mailing list
Matplotlib-users@lists.sourceforge.net
https://lists.sourceforge.net/lists/listinfo/matplotlib-users


Re: [Matplotlib-users] Runtime Error - Need Advice

2010-02-09 Thread Eric Firing
Lee Boger wrote:
 
 Thanks for the quick feedback. I changed the last line from 
 figure.figure.show() to pyplot.show() and it worked without crashing the 
 interpreter. However, if I close the figure then re-run the script, a 
 new figure pops up but it doesn't have any data plotted. Interpreter is 
 now locked up, the run-time error occurred and no figure stored as a 
 file log_plot. I un-installed numpy-1.4.0 and installed numpy-1.3.0, but 
 the results were as I just described.
 
 Lee
 


At this point a Windows user may need to step in.  I know nothing about 
PythonWin.  My suspicion is that this is a problem of dueling event 
loops, and one solution would be to use ipython instead of pythonwin.

(Note that you typically need to use reply to all on this list, 
otherwise the reply does not get copied to the list.)

Eric

 
 
 *Eric Firing efir...@hawaii.edu*
 
 02/09/2010 01:35 PM
 
   
 To
   Lee Boger boger_...@cat.com
 cc
   matplotlib-users@lists.sourceforge.net
 Subject
   Re: [Matplotlib-users] Runtime Error - Need Advice
 
 
   
 
 
 Caterpillar: Confidential GreenRetain Until: 03/11/2010
 
 
 
 
 Lee Boger wrote:
  
   Windows XP Professional with Python 2.5 installed (pywin32 build 210) -
   came with dSPACE software package
  
   Downloaded and installed matplotlib-0.99.1.win32-py2.5.exe from
   sourceforge.net
  
   Downloaded and installed numpy-1.4.0-win32-superpack-python2.5.exe from
   sourceforge.net
 
 I suspect this is the problem: numpy-1.4.0 is considered broken and has
 been withdrawn.  Specifically, it introduces binary incompatibility with
 programs compiled against prior versions of numpy, including matplotlib.
 If you go back to the numpy sourceforge site now, I think you will see
 an earlier version that you can install in place of 1.4.0 (after
 removing the latter).
 
 Eric
 
  
   Executing the following simple log plot script within PythonWin:
  
   *from* matplotlib *import* pylab
  
   # Create some artificial data.
   test_frequency = [1, 2, 3, 4, 5, 6, 7, 8, 9, 10, 11, 12, 13, 14, 15, 16,
   17, 18, 19, 20]
   test_results = [-0.2, -0.7, -1.0, -1.5, -2.0, -2.5, -3.0, -3.5, -4, -5,
   -6, -7.1, -8, -9, -10, -11, -12, -15, -20, -25]
   spec_frequency = [6, 8]
   spec_results = [-3.0, -3.0]
  
   # Plot
   figure = pylab.subplot(111)
   figure.semilogx()
   figure.scatter(test_frequency, test_results, s=20, c='b', marker='s',
   edgecolors='none')
   figure.scatter(spec_frequency, spec_results, s=40, c='g', marker='s',
   edgecolors='none')
   figure.grid(True)
   figure.set_xlabel(rFrequency (Hz), fontsize = 12)
   figure.set_ylabel(rActuator Response (db), fontsize = 12)
  
   figure.figure.savefig('log_plot')
   figure.figure.show()
  
  
   Plots a figure on the screen that looks correct, then the following
   error (when I click OK, PythonWin closes)
  
  
  
   Any advice would be appreciated. Maybe it's an installation or setup
   issue, but I'm pretty knew to Python programming and don't know how to
   debug this.
  
   Lee Boger
 


--
SOLARIS 10 is the OS for Data Centers - provides features such as DTrace,
Predictive Self Healing and Award Winning ZFS. Get Solaris 10 NOW
http://p.sf.net/sfu/solaris-dev2dev
___
Matplotlib-users mailing list
Matplotlib-users@lists.sourceforge.net
https://lists.sourceforge.net/lists/listinfo/matplotlib-users


Re: [Matplotlib-users] Runtime Error - Need Advice

2010-02-09 Thread Lee Boger
Thanks for your help. I'll look into ipython.

Lee



Eric Firing efir...@hawaii.edu 
02/09/2010 02:31 PM

To
Lee Boger boger_...@cat.com, matplotlib-users 
matplotlib-users@lists.sourceforge.net
cc

Subject
Re: [Matplotlib-users] Runtime Error - Need Advice




Caterpillar: Confidential Green Retain Until: 03/11/2010 



Lee Boger wrote:
 
 Thanks for the quick feedback. I changed the last line from 
 figure.figure.show() to pyplot.show() and it worked without crashing the 

 interpreter. However, if I close the figure then re-run the script, a 
 new figure pops up but it doesn't have any data plotted. Interpreter is 
 now locked up, the run-time error occurred and no figure stored as a 
 file log_plot. I un-installed numpy-1.4.0 and installed numpy-1.3.0, but 

 the results were as I just described.
 
 Lee
 


At this point a Windows user may need to step in.  I know nothing about 
PythonWin.  My suspicion is that this is a problem of dueling event 
loops, and one solution would be to use ipython instead of pythonwin.

(Note that you typically need to use reply to all on this list, 
otherwise the reply does not get copied to the list.)

Eric

 
 
 *Eric Firing efir...@hawaii.edu*
 
 02/09/2010 01:35 PM
 
 
 To
Lee Boger boger_...@cat.com
 cc
matplotlib-users@lists.sourceforge.net
 Subject
Re: [Matplotlib-users] Runtime Error - Need Advice
 
 
 
 
 
 Caterpillar: Confidential GreenRetain Until: 03/11/2010 
 
 
 
 
 Lee Boger wrote:
  
   Windows XP Professional with Python 2.5 installed (pywin32 build 210) 
-
   came with dSPACE software package
  
   Downloaded and installed matplotlib-0.99.1.win32-py2.5.exe from
   sourceforge.net
  
   Downloaded and installed numpy-1.4.0-win32-superpack-python2.5.exe 
from
   sourceforge.net
 
 I suspect this is the problem: numpy-1.4.0 is considered broken and has
 been withdrawn.  Specifically, it introduces binary incompatibility with
 programs compiled against prior versions of numpy, including matplotlib.
 If you go back to the numpy sourceforge site now, I think you will see
 an earlier version that you can install in place of 1.4.0 (after
 removing the latter).
 
 Eric
 
  
   Executing the following simple log plot script within PythonWin:
  
   *from* matplotlib *import* pylab
  
   # Create some artificial data.
   test_frequency = [1, 2, 3, 4, 5, 6, 7, 8, 9, 10, 11, 12, 13, 14, 15, 
16,
   17, 18, 19, 20]
   test_results = [-0.2, -0.7, -1.0, -1.5, -2.0, -2.5, -3.0, -3.5, -4, 
-5,
   -6, -7.1, -8, -9, -10, -11, -12, -15, -20, -25]
   spec_frequency = [6, 8]
   spec_results = [-3.0, -3.0]
  
   # Plot
   figure = pylab.subplot(111)
   figure.semilogx()
   figure.scatter(test_frequency, test_results, s=20, c='b', marker='s',
   edgecolors='none')
   figure.scatter(spec_frequency, spec_results, s=40, c='g', marker='s',
   edgecolors='none')
   figure.grid(True)
   figure.set_xlabel(rFrequency (Hz), fontsize = 12)
   figure.set_ylabel(rActuator Response (db), fontsize = 12)
  
   figure.figure.savefig('log_plot')
   figure.figure.show()
  
  
   Plots a figure on the screen that looks correct, then the following
   error (when I click OK, PythonWin closes)
  
  
  
   Any advice would be appreciated. Maybe it's an installation or setup
   issue, but I'm pretty knew to Python programming and don't know how 
to
   debug this.
  
   Lee Boger
 


--
SOLARIS 10 is the OS for Data Centers - provides features such as DTrace,
Predictive Self Healing and Award Winning ZFS. Get Solaris 10 NOW
http://p.sf.net/sfu/solaris-dev2dev___
Matplotlib-users mailing list
Matplotlib-users@lists.sourceforge.net
https://lists.sourceforge.net/lists/listinfo/matplotlib-users


Re: [Matplotlib-users] Runtime Error - Need Advice

2010-02-09 Thread Christoph Gohlke
I am not sure what's going on. The script you posted imports pylab in
the first line. It should be defined. The scrit works just fine after
the proposed changes, even from Pythonwin. Did you completely quit
Pythonwin after the crashes and made sure no corrupted python instance
was left running? Please send the version string of your python
installation.

Christoph

On 2/9/2010 11:10 AM, Lee Boger wrote:
 
 Per another suggestion, I did uninstall numpy 1.4 and installed numpy
 1.3, but got the same results. Using pylab.savefig('log_plot) resulted
 in the following error:
 
 Traceback (most recent call last):
   File C:\Program Files\Common
 Files\dSPACE\Python25\lib\site-packages\Pythonwin\pywin\framework\scriptutils.py,
 line 310, in RunScript
 exec codeObject in __main__.__dict__
   File C:\Documents and Settings\BogerLC\My Documents\Script2.py, line
 20, in module
 pylab.savefig('log_plot')
 NameError: name 'pylab' is not defined
 
 Using pyplot.savefig('log_plot') does not have an error, but I still get
 no file saved. Maybe it is a backend thing, which I know nothing
 about. I'll look into it. Thanks.
 
 Lee
 
 
 *Christoph Gohlke cgoh...@uci.edu*
 
 02/09/2010 02:03 PM
 
   
 To
   matplotlib-users matplotlib-users@lists.sourceforge.net
 cc
   
 Subject
   Re: [Matplotlib-users] Runtime Error - Need Advice
 
 
   
 
 
 Caterpillar: Confidential GreenRetain Until: 03/11/2010
 
 
 
 
 Use pylab.savefig('log_plot'). Also saving the figure to file and
 showing it interactively in the same script might involve switching
 backends, which might not work as expected.
 
 It is advisable to downgrade to numpy 1.3.
 
 -- Christoph
 
 On 2/9/2010 10:44 AM, Lee Boger wrote:

 Although, if I close the figure then re-run the script, a new figure
 pops up but it doesn't have any data plotted. Interpreter is now locked
 up. I'm still not fixing it completely. There is also no figure stored
 as a file log_plot.

 Lee


 *Christoph Gohlke cgoh...@uci.edu*

 02/09/2010 01:23 PM

  
 To
  Lee Boger boger_...@cat.com
 cc
  
 Subject
  Re: [Matplotlib-users] Runtime Error - Need Advice


  


 Caterpillar: Confidential GreenRetain Until: 03/11/2010




 Change the last line to pylab.show() and it should work.

 Anyway, this example should not crash the interpreter. I can reproduce
 the crash on Python 2.4, 2.5, and 2.6 (32 and 64 bit) on Windows with
 mpl 0.99.1 but not on Ubuntu 9.1 with mpl 0.99.0.

 The shortest example that crashes is:

 python -c import pylab;pylab.subplot(111).figure.show()

 or on the interactive prompt:

 import pylab
 pylab.subplot(111).figure.show()
 exit()
 Fatal Python error: PyEval_RestoreThread: NULL tstate

 This application has requested the Runtime to terminate it in an unusual
 way.
 Please contact the application's support team for more information.


 -- Christoph

 On 2/9/2010 7:57 AM, Lee Boger wrote:

 Windows XP Professional with Python 2.5 installed (pywin32 build 210) -
 came with dSPACE software package

 Downloaded and installed matplotlib-0.99.1.win32-py2.5.exe from
 sourceforge.net

 Downloaded and installed numpy-1.4.0-win32-superpack-python2.5.exe from
 sourceforge.net

 Executing the following simple log plot script within PythonWin:

 *from* matplotlib *import* pylab

 # Create some artificial data.
 test_frequency = [1, 2, 3, 4, 5, 6, 7, 8, 9, 10, 11, 12, 13, 14, 15, 16,
 17, 18, 19, 20]
 test_results = [-0.2, -0.7, -1.0, -1.5, -2.0, -2.5, -3.0, -3.5, -4, -5,
 -6, -7.1, -8, -9, -10, -11, -12, -15, -20, -25]
 spec_frequency = [6, 8]
 spec_results = [-3.0, -3.0]

 # Plot
 figure = pylab.subplot(111)
 figure.semilogx()
 figure.scatter(test_frequency, test_results, s=20, c='b', marker='s',
 edgecolors='none')
 figure.scatter(spec_frequency, spec_results, s=40, c='g', marker='s',
 edgecolors='none')
 figure.grid(True)
 figure.set_xlabel(rFrequency (Hz), fontsize = 12)
 figure.set_ylabel(rActuator Response (db), fontsize = 12)

 figure.figure.savefig('log_plot')
 figure.figure.show()


 Plots a figure on the screen that looks correct, then the following
 error (when I click OK, PythonWin closes)



 Any advice would be appreciated. Maybe it's an installation or setup
 issue, but I'm pretty knew to Python programming and don't know how to
 debug this.

 Lee Boger




 --
 The Planet: dedicated and managed hosting, cloud storage, colocation
 Stay online with enterprise data centers and the best network in the
 business
 Choose flexible plans and management services without long-term contracts
 Personal 24x7 support from experience hosting pros just a phone call away.
 http://p.sf.net/sfu/theplanet-com



 ___
 Matplotlib-users mailing list
 Matplotlib-users@lists.sourceforge.net
 

Re: [Matplotlib-users] Runtime Error - Need Advice

2010-02-09 Thread Lee Boger
I tried something that seems to work consistently - I get the correct plot 
figure and the correct log_plot.png file created everytime I run the 
script.

From the Start menu, I chose All Programs-Python2.5-IDLE (Python GUI) 
instead of All Programs-Python2.5-PythonWin, which creates the Tk Python 
Shell. I then open the script file and run it. I think you were right. 
Does this make sense?

Lee

__

Thanks for your help. I'll look into ipython.

Lee



Eric Firing efir...@hawaii.edu 
02/09/2010 02:31 PM

To
Lee Boger boger_...@cat.com, matplotlib-users 
matplotlib-users@lists.sourceforge.net
cc

Subject
Re: [Matplotlib-users] Runtime Error - Need Advice




Caterpillar: Confidential Green Retain Until: 03/11/2010 



Lee Boger wrote:
 
 Thanks for the quick feedback. I changed the last line from 
 figure.figure.show() to pyplot.show() and it worked without crashing the 

 interpreter. However, if I close the figure then re-run the script, a 
 new figure pops up but it doesn't have any data plotted. Interpreter is 
 now locked up, the run-time error occurred and no figure stored as a 
 file log_plot. I un-installed numpy-1.4.0 and installed numpy-1.3.0, but 

 the results were as I just described.
 
 Lee
 


At this point a Windows user may need to step in.  I know nothing about 
PythonWin.  My suspicion is that this is a problem of dueling event 
loops, and one solution would be to use ipython instead of pythonwin.

(Note that you typically need to use reply to all on this list, 
otherwise the reply does not get copied to the list.)

Eric

 
 
 *Eric Firing efir...@hawaii.edu*
 
 02/09/2010 01:35 PM
 
 
 To
Lee Boger boger_...@cat.com
 cc
matplotlib-users@lists.sourceforge.net
 Subject
Re: [Matplotlib-users] Runtime Error - Need Advice
 
 
 
 
 
 Caterpillar: Confidential GreenRetain Until: 03/11/2010 
 
 
 
 
 Lee Boger wrote:
  
   Windows XP Professional with Python 2.5 installed (pywin32 build 210) 
-
   came with dSPACE software package
  
   Downloaded and installed matplotlib-0.99.1.win32-py2.5.exe from
   sourceforge.net
  
   Downloaded and installed numpy-1.4.0-win32-superpack-python2.5.exe 
from
   sourceforge.net
 
 I suspect this is the problem: numpy-1.4.0 is considered broken and has
 been withdrawn.  Specifically, it introduces binary incompatibility with
 programs compiled against prior versions of numpy, including matplotlib.
 If you go back to the numpy sourceforge site now, I think you will see
 an earlier version that you can install in place of 1.4.0 (after
 removing the latter).
 
 Eric
 
  
   Executing the following simple log plot script within PythonWin:
  
   *from* matplotlib *import* pylab
  
   # Create some artificial data.
   test_frequency = [1, 2, 3, 4, 5, 6, 7, 8, 9, 10, 11, 12, 13, 14, 15, 
16,
   17, 18, 19, 20]
   test_results = [-0.2, -0.7, -1.0, -1.5, -2.0, -2.5, -3.0, -3.5, -4, 
-5,
   -6, -7.1, -8, -9, -10, -11, -12, -15, -20, -25]
   spec_frequency = [6, 8]
   spec_results = [-3.0, -3.0]
  
   # Plot
   figure = pylab.subplot(111)
   figure.semilogx()
   figure.scatter(test_frequency, test_results, s=20, c='b', marker='s',
   edgecolors='none')
   figure.scatter(spec_frequency, spec_results, s=40, c='g', marker='s',
   edgecolors='none')
   figure.grid(True)
   figure.set_xlabel(rFrequency (Hz), fontsize = 12)
   figure.set_ylabel(rActuator Response (db), fontsize = 12)
  
   figure.figure.savefig('log_plot')
   figure.figure.show()
  
  
   Plots a figure on the screen that looks correct, then the following
   error (when I click OK, PythonWin closes)
  
  
  
   Any advice would be appreciated. Maybe it's an installation or setup
   issue, but I'm pretty knew to Python programming and don't know how 
to
   debug this.
  
   Lee Boger
 


--
SOLARIS 10 is the OS for Data Centers - provides features such as DTrace,
Predictive Self Healing and Award Winning ZFS. Get Solaris 10 NOW
http://p.sf.net/sfu/solaris-dev2dev___
Matplotlib-users mailing list
Matplotlib-users@lists.sourceforge.net
https://lists.sourceforge.net/lists/listinfo/matplotlib-users


Re: [Matplotlib-users] Runtime Error - Need Advice

2010-02-09 Thread Lee Boger
Python version 2.5.1, Tk version 8.4, IDLE version 1.2.1

Lee



Christoph Gohlke cgoh...@uci.edu 
02/09/2010 03:08 PM

To
matplotlib-users matplotlib-users@lists.sourceforge.net
cc

Subject
Re: [Matplotlib-users] Runtime Error - Need Advice




Caterpillar: Confidential Green Retain Until: 03/11/2010 



I am not sure what's going on. The script you posted imports pylab in
the first line. It should be defined. The scrit works just fine after
the proposed changes, even from Pythonwin. Did you completely quit
Pythonwin after the crashes and made sure no corrupted python instance
was left running? Please send the version string of your python
installation.

Christoph

On 2/9/2010 11:10 AM, Lee Boger wrote:
 
 Per another suggestion, I did uninstall numpy 1.4 and installed numpy
 1.3, but got the same results. Using pylab.savefig('log_plot) resulted
 in the following error:
 
 Traceback (most recent call last):
   File C:\Program Files\Common
 
Files\dSPACE\Python25\lib\site-packages\Pythonwin\pywin\framework\scriptutils.py,
 line 310, in RunScript
 exec codeObject in __main__.__dict__
   File C:\Documents and Settings\BogerLC\My Documents\Script2.py, line
 20, in module
 pylab.savefig('log_plot')
 NameError: name 'pylab' is not defined
 
 Using pyplot.savefig('log_plot') does not have an error, but I still get
 no file saved. Maybe it is a backend thing, which I know nothing
 about. I'll look into it. Thanks.
 
 Lee
 
 
 *Christoph Gohlke cgoh...@uci.edu*
 
 02/09/2010 02:03 PM
 
 
 To
matplotlib-users matplotlib-users@lists.sourceforge.net
 cc
 
 Subject
Re: [Matplotlib-users] Runtime Error - Need Advice
 
 
 
 
 
 Caterpillar: Confidential GreenRetain Until: 03/11/2010 
 
 
 
 
 Use pylab.savefig('log_plot'). Also saving the figure to file and
 showing it interactively in the same script might involve switching
 backends, which might not work as expected.
 
 It is advisable to downgrade to numpy 1.3.
 
 -- Christoph
 
 On 2/9/2010 10:44 AM, Lee Boger wrote:

 Although, if I close the figure then re-run the script, a new figure
 pops up but it doesn't have any data plotted. Interpreter is now locked
 up. I'm still not fixing it completely. There is also no figure stored
 as a file log_plot.

 Lee


 *Christoph Gohlke cgoh...@uci.edu*

 02/09/2010 01:23 PM

 
 To
  Lee Boger boger_...@cat.com
 cc
 
 Subject
  Re: [Matplotlib-users] Runtime Error - Need Advice


 


 Caterpillar: Confidential GreenRetain Until: 03/11/2010 




 Change the last line to pylab.show() and it should work.

 Anyway, this example should not crash the interpreter. I can reproduce
 the crash on Python 2.4, 2.5, and 2.6 (32 and 64 bit) on Windows with
 mpl 0.99.1 but not on Ubuntu 9.1 with mpl 0.99.0.

 The shortest example that crashes is:

 python -c import pylab;pylab.subplot(111).figure.show()

 or on the interactive prompt:

 import pylab
 pylab.subplot(111).figure.show()
 exit()
 Fatal Python error: PyEval_RestoreThread: NULL tstate

 This application has requested the Runtime to terminate it in an 
unusual
 way.
 Please contact the application's support team for more information.


 -- Christoph

 On 2/9/2010 7:57 AM, Lee Boger wrote:

 Windows XP Professional with Python 2.5 installed (pywin32 build 210) 
-
 came with dSPACE software package

 Downloaded and installed matplotlib-0.99.1.win32-py2.5.exe from
 sourceforge.net

 Downloaded and installed numpy-1.4.0-win32-superpack-python2.5.exe 
from
 sourceforge.net

 Executing the following simple log plot script within PythonWin:

 *from* matplotlib *import* pylab

 # Create some artificial data.
 test_frequency = [1, 2, 3, 4, 5, 6, 7, 8, 9, 10, 11, 12, 13, 14, 15, 
16,
 17, 18, 19, 20]
 test_results = [-0.2, -0.7, -1.0, -1.5, -2.0, -2.5, -3.0, -3.5, -4, 
-5,
 -6, -7.1, -8, -9, -10, -11, -12, -15, -20, -25]
 spec_frequency = [6, 8]
 spec_results = [-3.0, -3.0]

 # Plot
 figure = pylab.subplot(111)
 figure.semilogx()
 figure.scatter(test_frequency, test_results, s=20, c='b', marker='s',
 edgecolors='none')
 figure.scatter(spec_frequency, spec_results, s=40, c='g', marker='s',
 edgecolors='none')
 figure.grid(True)
 figure.set_xlabel(rFrequency (Hz), fontsize = 12)
 figure.set_ylabel(rActuator Response (db), fontsize = 12)

 figure.figure.savefig('log_plot')
 figure.figure.show()


 Plots a figure on the screen that looks correct, then the following
 error (when I click OK, PythonWin closes)



 Any advice would be appreciated. Maybe it's an installation or setup
 issue, but I'm pretty knew to Python programming and don't know how to
 debug this.

 Lee Boger




 
--
 The Planet: dedicated and managed hosting, cloud storage, colocation
 Stay online with enterprise data centers and the best network in the
 business
 Choose flexible plans and management services without long-term 
contracts
 Personal 24x7 support from 

Re: [Matplotlib-users] Surface Plot

2010-02-09 Thread Ted Kord
Paul

The example works fine and was quite instructive. Thx.

Ted



On 8 February 2010 21:41, phob...@geosyntec.com wrote:

  Ted,



 How does this example run for you?


 http://matplotlib.sourceforge.net/examples/pylab_examples/griddata_demo.html



 *From:* Ted Kord [mailto:teddy.k...@googlemail.com]
 *Sent:* Monday, February 08, 2010 12:00 PM
 *To:* Paul Hobson
 *Cc:* matplotlib-users@lists.sourceforge.net
 *Subject:* Re: [Matplotlib-users] Surface Plot



 On 8 February 2010 17:37, phob...@geosyntec.com wrote:

 Hey Ted,

 I don't quite understand how you're getting the Z data below. But if you
 have 3D data in X, Y, and Z 1D-arrays, the griddata function should work for
 you.

 http://matplotlib.sourceforge.net/api/mlab_api.html#matplotlib.mlab.griddata

 HTH,
 -paul



 Hi  Paul



 Is there a way to colour the output? For some reason, the final output is
 in a single colour even though I've specified cm.jet for the colour map.



 Regards



 Ted



--
SOLARIS 10 is the OS for Data Centers - provides features such as DTrace,
Predictive Self Healing and Award Winning ZFS. Get Solaris 10 NOW
http://p.sf.net/sfu/solaris-dev2dev___
Matplotlib-users mailing list
Matplotlib-users@lists.sourceforge.net
https://lists.sourceforge.net/lists/listinfo/matplotlib-users


Re: [Matplotlib-users] Runtime Error - Need Advice

2010-02-09 Thread Christoph Gohlke
I can now reproduce this. It seems the same old problem that PythonWin
can not reliably run matplotlib because successive runs of the script
use the same interpreter. Ipython should work. Append pylab.close() to
your script; at least it will not crash on the second run.

-- Christoph

On 2/9/2010 10:44 AM, Lee Boger wrote:
 
 Although, if I close the figure then re-run the script, a new figure
 pops up but it doesn't have any data plotted. Interpreter is now locked
 up. I'm still not fixing it completely. There is also no figure stored
 as a file log_plot.
 
 Lee
 
 
 *Christoph Gohlke cgoh...@uci.edu*
 
 02/09/2010 01:23 PM
 
   
 To
   Lee Boger boger_...@cat.com
 cc
   
 Subject
   Re: [Matplotlib-users] Runtime Error - Need Advice
 
 
   
 
 
 Caterpillar: Confidential GreenRetain Until: 03/11/2010
 
 
 
 
 Change the last line to pylab.show() and it should work.
 
 Anyway, this example should not crash the interpreter. I can reproduce
 the crash on Python 2.4, 2.5, and 2.6 (32 and 64 bit) on Windows with
 mpl 0.99.1 but not on Ubuntu 9.1 with mpl 0.99.0.
 
 The shortest example that crashes is:
 
 python -c import pylab;pylab.subplot(111).figure.show()
 
 or on the interactive prompt:
 
 import pylab
 pylab.subplot(111).figure.show()
 exit()
 Fatal Python error: PyEval_RestoreThread: NULL tstate
 
 This application has requested the Runtime to terminate it in an unusual
 way.
 Please contact the application's support team for more information.
 
 
 -- Christoph
 
 On 2/9/2010 7:57 AM, Lee Boger wrote:

 Windows XP Professional with Python 2.5 installed (pywin32 build 210) -
 came with dSPACE software package

 Downloaded and installed matplotlib-0.99.1.win32-py2.5.exe from
 sourceforge.net

 Downloaded and installed numpy-1.4.0-win32-superpack-python2.5.exe from
 sourceforge.net

 Executing the following simple log plot script within PythonWin:

 *from* matplotlib *import* pylab

 # Create some artificial data.
 test_frequency = [1, 2, 3, 4, 5, 6, 7, 8, 9, 10, 11, 12, 13, 14, 15, 16,
 17, 18, 19, 20]
 test_results = [-0.2, -0.7, -1.0, -1.5, -2.0, -2.5, -3.0, -3.5, -4, -5,
 -6, -7.1, -8, -9, -10, -11, -12, -15, -20, -25]
 spec_frequency = [6, 8]
 spec_results = [-3.0, -3.0]

 # Plot
 figure = pylab.subplot(111)
 figure.semilogx()
 figure.scatter(test_frequency, test_results, s=20, c='b', marker='s',
 edgecolors='none')
 figure.scatter(spec_frequency, spec_results, s=40, c='g', marker='s',
 edgecolors='none')
 figure.grid(True)
 figure.set_xlabel(rFrequency (Hz), fontsize = 12)
 figure.set_ylabel(rActuator Response (db), fontsize = 12)

 figure.figure.savefig('log_plot')
 figure.figure.show()


 Plots a figure on the screen that looks correct, then the following
 error (when I click OK, PythonWin closes)



 Any advice would be appreciated. Maybe it's an installation or setup
 issue, but I'm pretty knew to Python programming and don't know how to
 debug this.

 Lee Boger



 --
 The Planet: dedicated and managed hosting, cloud storage, colocation
 Stay online with enterprise data centers and the best network in the
 business
 Choose flexible plans and management services without long-term contracts
 Personal 24x7 support from experience hosting pros just a phone call away.
 http://p.sf.net/sfu/theplanet-com



 ___
 Matplotlib-users mailing list
 Matplotlib-users@lists.sourceforge.net
 https://lists.sourceforge.net/lists/listinfo/matplotlib-users
 

--
SOLARIS 10 is the OS for Data Centers - provides features such as DTrace,
Predictive Self Healing and Award Winning ZFS. Get Solaris 10 NOW
http://p.sf.net/sfu/solaris-dev2dev
___
Matplotlib-users mailing list
Matplotlib-users@lists.sourceforge.net
https://lists.sourceforge.net/lists/listinfo/matplotlib-users


Re: [Matplotlib-users] hatching problem

2010-02-09 Thread Michiel de Hoon
I wasn't able to replicate this problem with the Mac OS X backend with 
matplotlib 0.99.1.1. Both the on-screen figure and the ps output look fine.

--Michiel.


--- On Tue, 2/9/10, Jae-Joon Lee lee.j.j...@gmail.com wrote:

 From: Jae-Joon Lee lee.j.j...@gmail.com
 Subject: Re: [Matplotlib-users] hatching problem
 To: Tomasz Koziara t.kozi...@civil.gla.ac.uk
 Cc: matplotlib-users@lists.sourceforge.net
 Date: Tuesday, February 9, 2010, 10:42 AM
 I cannot reproduce it with Agg
 backend and ps backend. I tried both
 svn version and 0.99 maint. version.
 So, maybe this is a bug in mac os X backend?
 Do you see a same problem with the ps output? If so, can
 you post your
 ouput ps file?
 
 Regards,
 
 -JJ
 
 
 On Mon, Feb 8, 2010 at 3:26 PM, Tomasz Koziara
 t.kozi...@civil.gla.ac.uk
 wrote:
  Dear Users/Developers
 
  I just installed version 0.99.1.1 since in my previous
 version (0.98) I had
  problems with hatching. It seems though that the same
 problems persist in
  the current version. The attached files reproduce the
 problem (a data file
  and a python short script). Note that hatching is not
 present on all green
  'CONUPD' fields - but only on few of them.
 
  I will appreciate some hints on how to get by,
  Regards
  Tomek
 
  ===
 
 
 
 
 
 
  ===
 
  dh178-192:tkp5 tomek$ python plots.py
 --verbose-helpful
  $HOME=/Users/tomek
  CONFIGDIR=/Users/tomek/.matplotlib
  matplotlib data path
 /Library/Python/2.5/site-packages/matplotlib/mpl-data
  loaded rc file
 
 /Library/Python/2.5/site-packages/matplotlib/mpl-data/matplotlibrc
  matplotlib version 0.99.1.1
  verbose.level helpful
  interactive is False
  units is False
  platform is darwin
  Using fontManager instance from
 /Users/tomek/.matplotlib/fontList.cache
  backend MacOSX version unknown
  findfont: Matching
 
 :family=sans-serif:style=normal:variant=normal:weight=normal:stretch=normal:size=medium
  to Bitstream Vera Sans
 
 (/Library/Python/2.5/site-packages/matplotlib-0.98.6svn-py2.5-macosx-10.3-fat.egg/matplotlib/mpl-data/fonts/ttf/Vera.ttf)
  with score of 0.00
  findfont: Matching
 
 :family=sans-serif:style=normal:variant=normal:weight=normal:stretch=normal:size=large
  to Bitstream Vera Sans
 
 (/Library/Python/2.5/site-packages/matplotlib-0.98.6svn-py2.5-macosx-10.3-fat.egg/matplotlib/mpl-data/fonts/ttf/Vera.ttf)
  with score of 0.00
 
 
 
 --
  The Planet: dedicated and managed hosting, cloud
 storage, colocation
  Stay online with enterprise data centers and the best
 network in the
  business
  Choose flexible plans and management services without
 long-term contracts
  Personal 24x7 support from experience hosting pros
 just a phone call away.
  http://p.sf.net/sfu/theplanet-com
  ___
  Matplotlib-users mailing list
  Matplotlib-users@lists.sourceforge.net
  https://lists.sourceforge.net/lists/listinfo/matplotlib-users
 
 
 
 --
 The Planet: dedicated and managed hosting, cloud storage,
 colocation
 Stay online with enterprise data centers and the best
 network in the business
 Choose flexible plans and management services without
 long-term contracts
 Personal 24x7 support from experience hosting pros just a
 phone call away.
 http://p.sf.net/sfu/theplanet-com
 ___
 Matplotlib-users mailing list
 Matplotlib-users@lists.sourceforge.net
 https://lists.sourceforge.net/lists/listinfo/matplotlib-users
 


  

--
SOLARIS 10 is the OS for Data Centers - provides features such as DTrace,
Predictive Self Healing and Award Winning ZFS. Get Solaris 10 NOW
http://p.sf.net/sfu/solaris-dev2dev
___
Matplotlib-users mailing list
Matplotlib-users@lists.sourceforge.net
https://lists.sourceforge.net/lists/listinfo/matplotlib-users


[Matplotlib-users] How Does One Learn to Use MatPlotLib?

2010-02-09 Thread Wayne Watson
Subject is the question.

As I see it, it's useful to know MatLab. A simple query with matplotlib 
tutorial shows a number of hits. The first, reference to v0.99.a 
documentation barely qualifies. Examples galore and a pretty minimal 
introduction. In the first 10 or so hits ther's a blog and mention of a 
video. The blog may appeal to some, but it seems unelementary. The video 
basically asks to sign in. Who knows where that goes? I've seen a few 
videos for MPL, but they all look tied into $$.

I've made some reasonable progress on MPL, but am still far short of 
being confident of using it.  Too much try this and see.

I know of exactly one book on MPL ( for scientists. sounds interesting). 
It was published recently by a foreign author. It is not yet widely 
distributed.

Your turn. Comments?
-- 
Crime is way down. War is declining. And that's far from the good 
news. -- Steven Pinker (and other sources) Why is this true, but yet 
the media says otherwise? The media knows very well how to manipulate us 
(see limbic, emotion, $$). -- WTW

--
SOLARIS 10 is the OS for Data Centers - provides features such as DTrace,
Predictive Self Healing and Award Winning ZFS. Get Solaris 10 NOW
http://p.sf.net/sfu/solaris-dev2dev
___
Matplotlib-users mailing list
Matplotlib-users@lists.sourceforge.net
https://lists.sourceforge.net/lists/listinfo/matplotlib-users


Re: [Matplotlib-users] How Does One Learn to Use MatPlotLib?

2010-02-09 Thread C M
sierra_mtnv...@sbcglobal.net wrote:
 Subject is the question.

 As I see it, it's useful to know MatLab. A simple query with matplotlib
 tutorial shows a number of hits. The first, reference to v0.99.a
 documentation barely qualifies. Examples galore and a pretty minimal
 introduction. In the first 10 or so hits ther's a blog and mention of a
 video. The blog may appeal to some, but it seems unelementary. The video
 basically asks to sign in. Who knows where that goes? I've seen a few
 videos for MPL, but they all look tied into $$.

 I've made some reasonable progress on MPL, but am still far short of
 being confident of using it.  Too much try this and see.

 I know of exactly one book on MPL ( for scientists. sounds interesting).
 It was published recently by a foreign author. It is not yet widely
 distributed.

 Your turn. Comments?

Have you read the manual itself?  I found that quite helpful in getting started:

http://matplotlib.sourceforge.net/users/index.html

I've been satisfied with learning a bit here and there as I go.  It's
a big project, and I don't expect to understand it all at once.  But,
that said, even with a partial understanding I'm pleased at what it
can do for my needs.  What I find is first you should determine what
is the desired outcome you want, then Google around for mentions of
this in old threads or on the matplotlib site, then maybe try asking
for assistance on the list.

Che

--
SOLARIS 10 is the OS for Data Centers - provides features such as DTrace,
Predictive Self Healing and Award Winning ZFS. Get Solaris 10 NOW
http://p.sf.net/sfu/solaris-dev2dev
___
Matplotlib-users mailing list
Matplotlib-users@lists.sourceforge.net
https://lists.sourceforge.net/lists/listinfo/matplotlib-users


Re: [Matplotlib-users] Runtime Error - Need Advice

2010-02-09 Thread Christoph Gohlke
To close this thread: the PyEval_RestoreThread crash is a known bug in
the tkagg backend on Windows. See line 375 of backend_tkagg.py:

def show(self):

this function doesn't segfault but causes the
PyEval_RestoreThread: NULL state bug on win32


-- Christoph

On 2/9/2010 10:26 AM, Christoph Gohlke wrote:
 Change the last line to pylab.show() and it should work.
 
 Anyway, this example should not crash the interpreter. I can reproduce
 the crash on Python 2.4, 2.5, and 2.6 (32 and 64 bit) on Windows with
 mpl 0.99.1 but not on Ubuntu 9.1 with mpl 0.99.0.
 
 The shortest example that crashes is:
 
 python -c import pylab;pylab.subplot(111).figure.show()
 
 or on the interactive prompt:
 
 import pylab
 pylab.subplot(111).figure.show()
 exit()
 Fatal Python error: PyEval_RestoreThread: NULL tstate
 
 This application has requested the Runtime to terminate it in an unusual
 way.
 Please contact the application's support team for more information.
 
 
 -- Christoph
 
 On 2/9/2010 7:57 AM, Lee Boger wrote:
  
  Windows XP Professional with Python 2.5 installed (pywin32 build 210) -
  came with dSPACE software package
  
  Downloaded and installed matplotlib-0.99.1.win32-py2.5.exe from
  sourceforge.net
  
  Downloaded and installed numpy-1.4.0-win32-superpack-python2.5.exe from
  sourceforge.net
  
  Executing the following simple log plot script within PythonWin:
  
  *from* matplotlib *import* pylab
  
  # Create some artificial data.
  test_frequency = [1, 2, 3, 4, 5, 6, 7, 8, 9, 10, 11, 12, 13, 14, 15, 16,
  17, 18, 19, 20]
  test_results = [-0.2, -0.7, -1.0, -1.5, -2.0, -2.5, -3.0, -3.5, -4, -5,
  -6, -7.1, -8, -9, -10, -11, -12, -15, -20, -25]
  spec_frequency = [6, 8]
  spec_results = [-3.0, -3.0]
  
  # Plot
  figure = pylab.subplot(111)
  figure.semilogx()
  figure.scatter(test_frequency, test_results, s=20, c='b', marker='s',
  edgecolors='none')
  figure.scatter(spec_frequency, spec_results, s=40, c='g', marker='s',
  edgecolors='none')
  figure.grid(True)
  figure.set_xlabel(rFrequency (Hz), fontsize = 12)
  figure.set_ylabel(rActuator Response (db), fontsize = 12)
  
  figure.figure.savefig('log_plot')
  figure.figure.show()
  
  
  Plots a figure on the screen that looks correct, then the following
  error (when I click OK, PythonWin closes)
  
  
  
  Any advice would be appreciated. Maybe it's an installation or setup
  issue, but I'm pretty knew to Python programming and don't know how to
  debug this.
  
  Lee Boger

--
SOLARIS 10 is the OS for Data Centers - provides features such as DTrace,
Predictive Self Healing and Award Winning ZFS. Get Solaris 10 NOW
http://p.sf.net/sfu/solaris-dev2dev
___
Matplotlib-users mailing list
Matplotlib-users@lists.sourceforge.net
https://lists.sourceforge.net/lists/listinfo/matplotlib-users


Re: [Matplotlib-users] How Does One Learn to Use MatPlotLib?

2010-02-09 Thread Gökhan Sever
On Tue, Feb 9, 2010 at 8:15 PM, Wayne Watson
sierra_mtnv...@sbcglobal.netwrote:

 Subject is the question.

 As I see it, it's useful to know MatLab. A simple query with matplotlib
 tutorial shows a number of hits. The first, reference to v0.99.a
 documentation barely qualifies. Examples galore and a pretty minimal
 introduction. In the first 10 or so hits ther's a blog and mention of a
 video. The blog may appeal to some, but it seems unelementary. The video
 basically asks to sign in. Who knows where that goes? I've seen a few
 videos for MPL, but they all look tied into $$.

 I've made some reasonable progress on MPL, but am still far short of
 being confident of using it.  Too much try this and see.

 I know of exactly one book on MPL ( for scientists. sounds interesting).
 It was published recently by a foreign author. It is not yet widely
 distributed.

 Your turn. Comments?
 --
 Crime is way down. War is declining. And that's far from the good
 news. -- Steven Pinker (and other sources) Why is this true, but yet
 the media says otherwise? The media knows very well how to manipulate us
 (see limbic, emotion, $$). -- WTW


 --
 SOLARIS 10 is the OS for Data Centers - provides features such as DTrace,
 Predictive Self Healing and Award Winning ZFS. Get Solaris 10 NOW
 http://p.sf.net/sfu/solaris-dev2dev
 ___
 Matplotlib-users mailing list
 Matplotlib-users@lists.sourceforge.net
 https://lists.sourceforge.net/lists/listinfo/matplotlib-users


For me best way to learn is to use it actually :) Especially on homework and
projects. Mailing lists are also very helpful as you are already doing.

Try with ipython --pylab option.

Also check SciPy09 (http://conference.scipy.org/SciPy2009/) videos. There
are one introductory and advanced tutorials that you can see online (without
registering) or downloading to your computer.



-- 
Gökhan
--
SOLARIS 10 is the OS for Data Centers - provides features such as DTrace,
Predictive Self Healing and Award Winning ZFS. Get Solaris 10 NOW
http://p.sf.net/sfu/solaris-dev2dev___
Matplotlib-users mailing list
Matplotlib-users@lists.sourceforge.net
https://lists.sourceforge.net/lists/listinfo/matplotlib-users