matplotlib-users Mailing List for matplotlib

On Tue, Sep 22, 2009 at 3:45 PM, Michael Hearne <mh...@us...> wrote:
> Great!

You may also want to take a look at the brand-spanking-new image
tutorial, which covers this

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

JDH

Great!

Thanks,

Mike
On Sep 22, 2009, at 2:38 PM, Gary Ruben wrote:

> Yes. Use interpolation='nearest' instead.
>
> Gary R.
>
> Michael Hearne wrote:
>> Running the test script below gives me the image I have attached,  
>> which looks like it has been smoothed.
>> Does imshow perform some sort of smoothing on the data it  
>> displays?  If so, is there a way to turn this off?
>> #!/usr/bin/env python
>> from pylab import *
>> data = array([[1,2,3,4],[5,6,7,8]])
>> imshow(data,interpolation=None)
>> savefig('output.png')
>> close('all')

Yes. Use interpolation='nearest' instead.

Gary R.

Michael Hearne wrote:
> Running the test script below gives me the image I have attached, which 
> looks like it has been smoothed.
> 
> Does imshow perform some sort of smoothing on the data it displays?  If 
> so, is there a way to turn this off?
> 
> #!/usr/bin/env python
> 
> from pylab import *
> 
> data = array([[1,2,3,4],[5,6,7,8]])
> imshow(data,interpolation=None)
> savefig('output.png')
> close('all')

Running the test script below gives me the image I have attached,  
which looks like it has been smoothed.

Does imshow perform some sort of smoothing on the data it displays?   
If so, is there a way to turn this off?

#!/usr/bin/env python

from pylab import *

data = array([[1,2,3,4],[5,6,7,8]])
imshow(data,interpolation=None)
savefig('output.png')
close('all')

On Tue, Sep 22, 2009 at 1:53 PM, Pierre GM <pgm...@gm...> wrote:
>
> On Sep 19, 2009, at 12:55 PM, Pierre GM wrote:
>
>> All,
>> I'm trying to use the Cocoa backend on Snow Leopard, using r7791
>> (GCC 4.2.1 / Python 2.6.1 from Apple, 64b)
>> Unfortunately, a simple `plot(range(10),range(10))` gives me an
>> empty window and error message as such:
>>
>> Python[53010:d07] Inconsistent set of values to create
>> NSBitmapImageRep
>> /Users/pierregm/.local/lib/python2.6/site-packages/matplotlib/
>> backends/backend_cocoaagg.py:140: UninitializedDeallocWarning:
>> leaking an uninitialized object of type NSBitmapImageRep
>>  32) # bits per pixel
>>
>> Can anybody reproduce it ? Would anybody have some ideas on how to
>> fix that ?
>> Thx a lot in advance.
>> P.
>> FYI, the build log can be accessed at: http://pastebin.com/d5b3c1838
>
> I'm very, very sorry to bump my own thread, but I'm in a bit of a
> pickle. Matplotlib installs and runs nicely on my Snow Leopard
> installation, but I'm "blind": cocoa fails, wxpython is 32b only and
> compiling Qt4 is becoming increasingly frustrating on a 64b machine.
> Any help or hint would be really, really appreciated.
> (and once again, sorry for my being rude)

Are you using the CocoaAgg backend or MacOSX?  If the former, is the
latter an option for you?

JDH

On Tue, Sep 22, 2009 at 2:53 PM, Pierre GM <pgm...@gm...> wrote:
>
> On Sep 19, 2009, at 12:55 PM, Pierre GM wrote:
>
>> All,
>> I'm trying to use the Cocoa backend on Snow Leopard, using r7791
>> (GCC 4.2.1 / Python 2.6.1 from Apple, 64b)
>> Unfortunately, a simple `plot(range(10),range(10))` gives me an
>> empty window and error message as such:
>>
>> Python[53010:d07] Inconsistent set of values to create
>> NSBitmapImageRep
>> /Users/pierregm/.local/lib/python2.6/site-packages/matplotlib/
>> backends/backend_cocoaagg.py:140: UninitializedDeallocWarning:
>> leaking an uninitialized object of type NSBitmapImageRep
>>  32) # bits per pixel
>>
>> Can anybody reproduce it ? Would anybody have some ideas on how to
>> fix that ?
>> Thx a lot in advance.
>> P.
>> FYI, the build log can be accessed at: http://pastebin.com/d5b3c1838
>
> I'm very, very sorry to bump my own thread, but I'm in a bit of a
> pickle. Matplotlib installs and runs nicely on my Snow Leopard
> installation, but I'm "blind": cocoa fails, wxpython is 32b only and
> compiling Qt4 is becoming increasingly frustrating on a 64b machine.
> Any help or hint would be really, really appreciated.
> (and once again, sorry for my being rude)
> P.

I'm not sure if you have the option to go 32-bit, but the cocoa
backend works fine on snow leopard with a 32-bit build (-arch i386),
that I built on leopard prior to upgrading. The 64 bit transition
should be interesting, especially given the persistence of heritage
universal (PPC) build processes.

-Eric

On Sep 19, 2009, at 12:55 PM, Pierre GM wrote:

> All,
> I'm trying to use the Cocoa backend on Snow Leopard, using r7791  
> (GCC 4.2.1 / Python 2.6.1 from Apple, 64b)
> Unfortunately, a simple `plot(range(10),range(10))` gives me an  
> empty window and error message as such:
>
> Python[53010:d07] Inconsistent set of values to create  
> NSBitmapImageRep
> /Users/pierregm/.local/lib/python2.6/site-packages/matplotlib/ 
> backends/backend_cocoaagg.py:140: UninitializedDeallocWarning:  
> leaking an uninitialized object of type NSBitmapImageRep
>  32) # bits per pixel
>
> Can anybody reproduce it ? Would anybody have some ideas on how to  
> fix that ?
> Thx a lot in advance.
> P.
> FYI, the build log can be accessed at: http://pastebin.com/d5b3c1838

I'm very, very sorry to bump my own thread, but I'm in a bit of a  
pickle. Matplotlib installs and runs nicely on my Snow Leopard  
installation, but I'm "blind": cocoa fails, wxpython is 32b only and  
compiling Qt4 is becoming increasingly frustrating on a 64b machine.
Any help or hint would be really, really appreciated.
(and once again, sorry for my being rude)
P.

John [H2O] wrote:
> Hello,
> 
> I'm trying to plot text objects with the color used in m.scatter. My goal is
> to individually label each scatter object. I've tried the following:
> 
> circles=m.scatter(cx,cy,zsize,zlevel,cmap=cmap)
> 
> and then:
> CC = circles.get_facecolor()
> i=0
> for x,y in map(cx,cy):
>     ax.text(x,y,'some text', color=CC[i])
> 
> But if fails because:
> 
> m.scatter returns:
> <matplotlib.collections.CircleCollection object at 0x127898d0>
> 
> Rather than a list of objects??? Or something. The point is, I would expect
> 'circles' above to have len==cx, but it does not. What am I doing wrong?

The first thing you are doing wrong is failing to provide a complete, 
minimal, self-contained example illustrating the problem.  I don't think 
the snippet you gave is representative of what you are running.  Instead 
of "map" I think you mean "zip".  Where are you incrementing "i"?

The general approach of using the get_facecolor() method to get the 
array, and indexing into it, is fine.  That is not where the problem is. 
  Once you put together a minimal example, I think you will find the 
problem--or that there really is no problem.

A collection is used by scatter because it is much more efficient than a 
sequence of objects.

Eric

> 
> 
> 

John,

The following code works for me (Python 2.5.4, Matplotlib 0.99):

# create and format the colorbar
cbar = pl.colorbar(G, ticks=range(g1,g2+1))    
cbar.ax.set_ylabel('Gradient (%)', fontsize=10)
cl = pl.getp(cbar.ax, 'ymajorticklabels')
pl.setp(cl, fontsize=10)

-Paul M. Hobson                  


> -----Original Message-----
> From: John [H2O] [mailto:was...@gm...]
> Sent: Tuesday, September 22, 2009 8:03 AM
> To: mat...@li...
> Subject: [Matplotlib-users] colorbar tick label fontproperties
> 
> 
> Could someone please explain how to change the font properties for a
> colorbar
> tick label, the following causes no error, but it does NOT work:
> 
>     ## CREATE COLORBAR
>     ## make a copy of the image object
>     im2 = copy.copy(im)
>     im2.set_cmap(colmap)
>     ## create new axis for colorbar.
>     cax = plt.axes([l+w+0.03, b, 0.025, h-0.035])
>     cb = plt.colorbar(im2, cax)#, format='%3.2g') # draw colorbar
>     ## set colorbar label and ticks
>     p_cax = mpl.font_manager.FontProperties(size='6')
>     clabels = clevs[::10] ##clevs, by 10 steps
>     clabels.append(clevs[-1]) ## add the last label
>     cb.ax.set_yticks(np.linspace(0,1,len(clabels)))
>     cb.ax.set_yticklabels(['%3.2g' % cl for cl in clabels],
>                         fontproperties=p_cax)
>     cax.set_title('sensitivity\n(%s)' % units,
>                       fontproperties=p_cax)
> 
> 
> NOTE: It does set the title properties correctly.
> --
> View this message in context: http://www.nabble.com/colorbar-tick-
> label-fontproperties-tp25530779p25530779.html
> Sent from the matplotlib - users mailing list archive at Nabble.com.
> 
> 
> ----------------------------------------------------------------------
> --------
> Come build with us! The BlackBerry® Developer Conference in SF, CA
> is the only developer event you need to attend this year. Jumpstart
> your
> developing skills, take BlackBerry mobile applications to market and
> stay
> ahead of the curve. Join us from November 9-12, 2009. Register
> now!
> http://p.sf.net/sfu/devconf
> _______________________________________________
> Matplotlib-users mailing list
> Mat...@li...
> https://lists.sourceforge.net/lists/listinfo/matplotlib-users

Hello,

I'm trying to plot text objects with the color used in m.scatter. My goal is
to individually label each scatter object. I've tried the following:

circles=m.scatter(cx,cy,zsize,zlevel,cmap=cmap)

and then:
CC = circles.get_facecolor()
i=0
for x,y in map(cx,cy):
    ax.text(x,y,'some text', color=CC[i])

But if fails because:

m.scatter returns:
<matplotlib.collections.CircleCollection object at 0x127898d0>

Rather than a list of objects??? Or something. The point is, I would expect
'circles' above to have len==cx, but it does not. What am I doing wrong?



-- 
View this message in context: http://www.nabble.com/circle-collection-get_facecolors-in-Basemap-tp25530811p25530811.html
Sent from the matplotlib - users mailing list archive at Nabble.com.

On Tue, Sep 22, 2009 at 10:25 AM, Alan <ala...@gm...> wrote:

> Hi there,
>
> I want to move to numpy only. What would be the replacement for:
>
> from matplotlib.numerix.mlab import amax
>
> ?
>

numpy.amax

You should be able to do that for most of numerix I believe (I and would
definitely try that first).

Ryan

-- 
Ryan May
Graduate Research Assistant
School of Meteorology
University of Oklahoma

Hi there,

I want to move to numpy only. What would be the replacement for:

from matplotlib.numerix.mlab import amax

?

Many thanks in advance,

Alan

Could someone please explain how to change the font properties for a colorbar
tick label, the following causes no error, but it does NOT work:

    ## CREATE COLORBAR 
    ## make a copy of the image object
    im2 = copy.copy(im)
    im2.set_cmap(colmap)
    ## create new axis for colorbar.
    cax = plt.axes([l+w+0.03, b, 0.025, h-0.035])
    cb = plt.colorbar(im2, cax)#, format='%3.2g') # draw colorbar
    ## set colorbar label and ticks
    p_cax = mpl.font_manager.FontProperties(size='6')
    clabels = clevs[::10] ##clevs, by 10 steps
    clabels.append(clevs[-1]) ## add the last label
    cb.ax.set_yticks(np.linspace(0,1,len(clabels)))
    cb.ax.set_yticklabels(['%3.2g' % cl for cl in clabels],
                        fontproperties=p_cax)
    cax.set_title('sensitivity\n(%s)' % units,
                      fontproperties=p_cax)


NOTE: It does set the title properties correctly.
-- 
View this message in context: http://www.nabble.com/colorbar-tick-label-fontproperties-tp25530779p25530779.html
Sent from the matplotlib - users mailing list archive at Nabble.com.

On Tue, Sep 22, 2009 at 9:22 AM, Alan <ala...@gm...> wrote:

> Hi there,
>
> In mtpl __version__  = '0.98.5.3' "from matplotlib._cm import LUTSIZE"
> still works, but in matplotlib.__version__ = '0.99.0' no:
>
> from matplotlib._cm import LUTSIZE
> ImportError: cannot import name LUTSIZE
>
> How can I replace/solve this issue?
>
> Many thanks in advance.
>

from matplotlib.cm import LUTSIZE

Ryan

-- 
Ryan May
Graduate Research Assistant
School of Meteorology
University of Oklahoma

Hi there,

In mtpl __version__  = '0.98.5.3' "from matplotlib._cm import LUTSIZE"
still works, but in matplotlib.__version__ = '0.99.0' no:

from matplotlib._cm import LUTSIZE
ImportError: cannot import name LUTSIZE

How can I replace/solve this issue?

Many thanks in advance.

Alan

The 0.99.1 release is available for download from sourceforge

  http://sourceforge.net/projects/matplotlib/files/matplotlib/matplotlib-0.99.1/

This is a bug-fix release with an emphasis on stability.  Lots of bugs
have been fixed since 0.99.0 so thanks to all who contributed bug
reports, patches and fixes.  A detailed listing of fixes is available
in the svn log below

Thanks to Christoph Gohlke for the windows builds.



------------------------------------------------------------------------
r7813 | jdh2358 | 2009-09-21 12:12:47 -0500 (Mon, 21 Sep 2009) | 1 line

tag for 0.99.1 release
------------------------------------------------------------------------
r7808 | ryanmay | 2009-09-21 11:30:11 -0500 (Mon, 21 Sep 2009) | 1 line

Correct typo of subtitle() -> suptitle().
------------------------------------------------------------------------
r7803 | mdboom | 2009-09-21 06:57:17 -0500 (Mon, 21 Sep 2009) | 1 line

Fix 'variable used before defined' errors in backend_pdf.py building the docs
------------------------------------------------------------------------
r7800 | jouni | 2009-09-20 14:47:46 -0500 (Sun, 20 Sep 2009) | 1 line

Fix usetex spacing errors in pdf backend.
------------------------------------------------------------------------
r7796 | jouni | 2009-09-20 08:19:25 -0500 (Sun, 20 Sep 2009) | 1 line

Fix off-by-one error in dviread.Tfm
------------------------------------------------------------------------
r7794 | jouni | 2009-09-20 07:30:22 -0500 (Sun, 20 Sep 2009) | 1 line

Prevent exception in case of missing height and depth information in a TeX font
------------------------------------------------------------------------
r7792 | efiring | 2009-09-19 18:46:06 -0500 (Sat, 19 Sep 2009) | 2 lines

Fix bug in bar affecting autoscaling with log axis

------------------------------------------------------------------------
r7790 | efiring | 2009-09-19 01:28:43 -0500 (Sat, 19 Sep 2009) | 2 lines

Fix bug in mlab.demean, with axis = -1

------------------------------------------------------------------------
r7788 | efiring | 2009-09-18 21:53:06 -0500 (Fri, 18 Sep 2009) | 2 lines

Fix bug in maxdict when replacing an existing item

------------------------------------------------------------------------
r7784 | mdboom | 2009-09-18 10:16:20 -0500 (Fri, 18 Sep 2009) | 2 lines

Fix bug where subslicing was cutting polar lines off (because it was
treating them as if they were rectilinearly plotted).

------------------------------------------------------------------------
r7780 | efiring | 2009-09-17 21:48:03 -0500 (Thu, 17 Sep 2009) | 2 lines

Fix garbled getp output

------------------------------------------------------------------------
r7778 | leejjoon | 2009-09-17 16:42:03 -0500 (Thu, 17 Sep 2009) | 1 line

fix the legend bug that dash-style for LineCollections handle is set incorrectly
------------------------------------------------------------------------
r7776 | jdh2358 | 2009-09-17 12:06:02 -0500 (Thu, 17 Sep 2009) | 1 line

add make.osx to manifest
------------------------------------------------------------------------
r7774 | mdboom | 2009-09-17 09:56:00 -0500 (Thu, 17 Sep 2009) | 2 lines

Fix bug whereby non-finite values in the path were not being ignored
by the view limits finding algorithm.

------------------------------------------------------------------------
r7772 | mdboom | 2009-09-17 09:17:48 -0500 (Thu, 17 Sep 2009) | 2 lines

Fix clip path in SVG backend.

------------------------------------------------------------------------
r7770 | efiring | 2009-09-16 18:38:54 -0500 (Wed, 16 Sep 2009) | 2 lines

Don't use wxversion with py2exe; fix by Werner Bruhin

------------------------------------------------------------------------
r7762 | jouni | 2009-09-15 07:12:19 -0500 (Tue, 15 Sep 2009) | 1 line

Don't fail on AFM files containing floating-point bounding boxes
------------------------------------------------------------------------
r7756 | mdboom | 2009-09-14 12:11:06 -0500 (Mon, 14 Sep 2009) | 1 line

Minor doc fixes
------------------------------------------------------------------------
r7751 | jdh2358 | 2009-09-12 16:11:42 -0500 (Sat, 12 Sep 2009) | 1 line

tagging for 99.1 release candiate
------------------------------------------------------------------------
r7749 | efiring | 2009-09-11 16:14:15 -0500 (Fri, 11 Sep 2009) | 2 lines

Backported quiver bug fix from trunk

------------------------------------------------------------------------
r7745 | efiring | 2009-09-11 15:48:10 -0500 (Fri, 11 Sep 2009) | 2 lines

Fix bug in quiver angle kwarg, found when input angle array is not 1-D

------------------------------------------------------------------------
r7741 | jdh2358 | 2009-09-10 21:06:30 -0500 (Thu, 10 Sep 2009) | 1 line

minor tweaks to licensing devel doc
------------------------------------------------------------------------
r7740 | astraw | 2009-09-10 20:54:27 -0500 (Thu, 10 Sep 2009) | 2 lines

bugfix: bounds checking in hexbin with extent specified (SF#2856228)

------------------------------------------------------------------------
r7734 | leejjoon | 2009-09-10 18:01:28 -0500 (Thu, 10 Sep 2009) | 1 line

fix a bug in Line2D.draw method that produces invalid svg when the
line is invisible
------------------------------------------------------------------------
r7728 | mdboom | 2009-09-09 14:55:52 -0500 (Wed, 09 Sep 2009) | 2 lines

Fix some documentation warnings.

------------------------------------------------------------------------
r7727 | ryanmay | 2009-09-09 13:41:43 -0500 (Wed, 09 Sep 2009) | 1 line

Include 'top' keyword argument in docstring for Figure.subplots_adjust().
------------------------------------------------------------------------
r7703 | leejjoon | 2009-09-07 17:29:04 -0500 (Mon, 07 Sep 2009) | 1 line

fix a bug in lib/matplotlib/bezier.py
------------------------------------------------------------------------
r7638 | astraw | 2009-09-05 18:20:45 -0500 (Sat, 05 Sep 2009) | 4 lines

spines: fix 'axes' position bug (closes SF#2852168)

Thanks to Jason Grout for reporting this.

------------------------------------------------------------------------
r7633 | leejjoon | 2009-09-03 22:35:25 -0500 (Thu, 03 Sep 2009) | 1 line

fix a bug in axes_grid.inset_locator.inset_axes
------------------------------------------------------------------------
r7618 | ryanmay | 2009-08-31 12:34:50 -0500 (Mon, 31 Aug 2009) | 1 line

Use atleast_1d instead of asanyarray.  This allows passing in scalars
to quiver/barbs.
------------------------------------------------------------------------
r7617 | ryanmay | 2009-08-31 12:29:41 -0500 (Mon, 31 Aug 2009) | 1 line

Pull _parse_args out of Quiver and Barbs classes, as we have multiple,
diverging copies of the same code.  Now the shared code will actually
be shared.
------------------------------------------------------------------------
r7616 | ryanmay | 2009-08-31 12:27:12 -0500 (Mon, 31 Aug 2009) | 1 line

Update barb_demo.py with an example using masked arrays.
------------------------------------------------------------------------
r7607 | leejjoon | 2009-08-30 19:32:12 -0500 (Sun, 30 Aug 2009) | 1 line

fix sf bug #2839919
------------------------------------------------------------------------
r7600 | jdh2358 | 2009-08-30 11:22:38 -0500 (Sun, 30 Aug 2009) | 1 line

some unit cleanup; fix sf bug 2846058
------------------------------------------------------------------------
r7598 | jdh2358 | 2009-08-30 08:35:12 -0500 (Sun, 30 Aug 2009) | 1 line

applied Gael's ginput patch
------------------------------------------------------------------------
r7589 | jdh2358 | 2009-08-29 22:40:08 -0500 (Sat, 29 Aug 2009) | 1 line

added minor kwarg to get_?ticklines and applied ginpu patch
------------------------------------------------------------------------
r7584 | efiring | 2009-08-29 12:48:37 -0500 (Sat, 29 Aug 2009) | 2 lines

Semilogx, semilogy ensure one axis is linear; patch by Tony Yu

------------------------------------------------------------------------
r7583 | jdh2358 | 2009-08-29 11:46:44 -0500 (Sat, 29 Aug 2009) | 1 line

minor tweaks to the image tut
------------------------------------------------------------------------
r7582 | jdh2358 | 2009-08-29 11:35:35 -0500 (Sat, 29 Aug 2009) | 1 line

added Michael Sarahan's image tutorial from the scipy mpl sprint
------------------------------------------------------------------------
r7569 | mdboom | 2009-08-25 15:04:34 -0500 (Tue, 25 Aug 2009) | 2 lines

Fix cohere_pairs docstring so it builds with new versions of Sphinx

------------------------------------------------------------------------
r7567 | mdboom | 2009-08-25 10:31:10 -0500 (Tue, 25 Aug 2009) | 2 lines

Support Line2D objects without associated Axes objects.

------------------------------------------------------------------------
r7541 | jdh2358 | 2009-08-23 13:42:37 -0500 (Sun, 23 Aug 2009) | 1 line

more harness around locale
------------------------------------------------------------------------
r7536 | jdh2358 | 2009-08-23 00:27:40 -0500 (Sun, 23 Aug 2009) | 1 line

fix some typos in the docs
------------------------------------------------------------------------
r7533 | jdh2358 | 2009-08-22 21:27:44 -0500 (Sat, 22 Aug 2009) | 1 line

tweaks to os x
------------------------------------------------------------------------
r7530 | jdh2358 | 2009-08-22 21:10:59 -0500 (Sat, 22 Aug 2009) | 1 line

removed darwin lookup in setupext -- use make.osx install
------------------------------------------------------------------------
r7529 | jdh2358 | 2009-08-22 21:10:46 -0500 (Sat, 22 Aug 2009) | 1 line

removed darwin lookup in setupext -- use make.osx install
------------------------------------------------------------------------
r7528 | jdh2358 | 2009-08-22 20:50:21 -0500 (Sat, 22 Aug 2009) | 1 line

added new makefile for osx builds
------------------------------------------------------------------------
r7523 | jdh2358 | 2009-08-22 18:20:30 -0500 (Sat, 22 Aug 2009) | 1 line

fix osx epd formatting bug in rest
------------------------------------------------------------------------
r7522 | jdh2358 | 2009-08-22 18:19:44 -0500 (Sat, 22 Aug 2009) | 1 line

applied Ariel's mlab.cohere_pairs fixes
------------------------------------------------------------------------
r7521 | jdh2358 | 2009-08-22 17:50:55 -0500 (Sat, 22 Aug 2009) | 1 line

added Ariel's max install patch
------------------------------------------------------------------------
r7519 | jouni | 2009-08-22 01:25:07 -0500 (Sat, 22 Aug 2009) | 2 lines

Fix typos found by Marc Desmarais and Nicolas Pinto

------------------------------------------------------------------------
r7517 | jdh2358 | 2009-08-21 19:02:15 -0500 (Fri, 21 Aug 2009) | 1 line

fixed a fifo bug for the new transforms infrastructure
------------------------------------------------------------------------
r7506 | ryanmay | 2009-08-19 02:56:33 -0500 (Wed, 19 Aug 2009) | 1 line

Remove calls to np.asarray().  This was breaking the use of masked
arrays in calls to set_[x|y]data() and is handled appropriately
already by set_data().
------------------------------------------------------------------------
r7502 | mdboom | 2009-08-18 09:08:27 -0500 (Tue, 18 Aug 2009) | 2 lines

Fix some minor typos in the paths tutorial

------------------------------------------------------------------------
r7501 | mdboom | 2009-08-18 09:01:41 -0500 (Tue, 18 Aug 2009) | 2 lines

Fix some minor typos in the transforms tutorial

------------------------------------------------------------------------
r7500 | mdboom | 2009-08-18 09:00:28 -0500 (Tue, 18 Aug 2009) | 2 lines

Fix some minor typos in the transforms tutorial

------------------------------------------------------------------------
r7499 | jdh2358 | 2009-08-18 00:15:21 -0500 (Tue, 18 Aug 2009) | 1 line

added looking glass demo
------------------------------------------------------------------------
r7498 | jdh2358 | 2009-08-17 22:48:21 -0500 (Mon, 17 Aug 2009) | 1 line

added path tut
------------------------------------------------------------------------
r7490 | jdh2358 | 2009-08-15 12:56:44 -0500 (Sat, 15 Aug 2009) | 1 line

added transformations tut, did some reorg
------------------------------------------------------------------------
r7489 | jdh2358 | 2009-08-15 12:40:27 -0500 (Sat, 15 Aug 2009) | 1 line

some doc reorg
------------------------------------------------------------------------
r7486 | mdboom | 2009-08-14 08:30:32 -0500 (Fri, 14 Aug 2009) | 2 lines

Fix documentation about Axis.transAxis (thanks Jason Sage)

------------------------------------------------------------------------
r7484 | efiring | 2009-08-13 20:58:44 -0500 (Thu, 13 Aug 2009) | 3 lines

Prevent exception when image is off screen and out of the axes.
Bug reported by G. Jones.

------------------------------------------------------------------------
r7482 | efiring | 2009-08-13 19:24:40 -0500 (Thu, 13 Aug 2009) | 2 lines

Remove older versions of some functions in mlab.py; closes bug 2806535

------------------------------------------------------------------------
r7480 | jdh2358 | 2009-08-12 19:41:58 -0500 (Wed, 12 Aug 2009) | 1 line

add the pngs referenced by sphinx css; closes sf bug 2834121
------------------------------------------------------------------------
r7477 | efiring | 2009-08-12 11:53:41 -0500 (Wed, 12 Aug 2009) | 2 lines

Typo in navigation_toolbar.rst.

------------------------------------------------------------------------
r7476 | jdh2358 | 2009-08-12 06:37:13 -0500 (Wed, 12 Aug 2009) | 1 line

do case insensitive color string matching, as suggested in sf bug 2834598
------------------------------------------------------------------------
r7475 | jdh2358 | 2009-08-11 19:26:50 -0500 (Tue, 11 Aug 2009) | 1 line

update the contributing faq
------------------------------------------------------------------------
r7446 | heeres | 2009-08-11 02:06:15 -0500 (Tue, 11 Aug 2009) | 2 lines

Fix bugs: #2830483 (axis scaling), 2834105 (z order problem)

------------------------------------------------------------------------
r7444 | efiring | 2009-08-09 19:31:30 -0500 (Sun, 09 Aug 2009) | 3 lines

Sage patch, slightly modified, to improve tk/tcl detection.
http://trac.sagemath.org/sage_trac/ticket/4176

------------------------------------------------------------------------
r7443 | efiring | 2009-08-09 19:19:19 -0500 (Sun, 09 Aug 2009) | 3 lines

Patch from Jason, sage project: prevent failure with tiny arrows.
Also deleting unwanted whitespace in bezier.py.

------------------------------------------------------------------------
r7442 | efiring | 2009-08-09 19:02:56 -0500 (Sun, 09 Aug 2009) | 2 lines

patch from sage project: make unicode_safe safer

------------------------------------------------------------------------
r7433 | efiring | 2009-08-08 13:16:01 -0500 (Sat, 08 Aug 2009) | 2 lines

Fix excessive line length in annotations.rst

------------------------------------------------------------------------
r7432 | jdh2358 | 2009-08-08 10:16:57 -0500 (Sat, 08 Aug 2009) | 1 line

simplify poly array in example
------------------------------------------------------------------------
r7431 | jdh2358 | 2009-08-08 10:10:08 -0500 (Sat, 08 Aug 2009) | 1 line

use a class helper method to make the compound path from polys
------------------------------------------------------------------------
r7430 | jdh2358 | 2009-08-08 08:58:52 -0500 (Sat, 08 Aug 2009) | 1 line

two new examples using a compund path for a histogram; one animated
------------------------------------------------------------------------
r7429 | jdh2358 | 2009-08-08 08:53:24 -0500 (Sat, 08 Aug 2009) | 1 line

two new examples using a compund path for a histogram; one animated
------------------------------------------------------------------------
r7428 | jdh2358 | 2009-08-08 07:21:29 -0500 (Sat, 08 Aug 2009) | 1 line

clean up mplot3d examples: use pyplot noy pylab and numpy rather than
list comps and python random module
------------------------------------------------------------------------
r7426 | jdh2358 | 2009-08-08 06:00:41 -0500 (Sat, 08 Aug 2009) | 1 line

replace list comps w/ numpy in mplot3d
------------------------------------------------------------------------
r7424 | efiring | 2009-08-08 01:07:06 -0500 (Sat, 08 Aug 2009) | 2 lines

Restore default colormap behavior: no color (alpha = 0) for masked data

------------------------------------------------------------------------
r7416 | jdh2358 | 2009-08-07 10:40:56 -0500 (Fri, 07 Aug 2009) | 1 line

don't fail on window icon load
------------------------------------------------------------------------
r7414 | jdh2358 | 2009-08-07 05:15:04 -0500 (Fri, 07 Aug 2009) | 1 line

some doc fixes
------------------------------------------------------------------------
r7409 | ryanmay | 2009-08-06 14:28:16 -0500 (Thu, 06 Aug 2009) | 1 line

Tweak solution for hiding colorbar_doc to sync with trunk.
------------------------------------------------------------------------
r7407 | jdh2358 | 2009-08-06 13:51:58 -0500 (Thu, 06 Aug 2009) | 1 line

hide colorbar_doc a bit
------------------------------------------------------------------------
r7404 | jdh2358 | 2009-08-06 11:41:18 -0500 (Thu, 06 Aug 2009) | 1 line

some updates to site docs for release
------------------------------------------------------------------------
r7403 | leejjoon | 2009-08-06 11:11:30 -0500 (Thu, 06 Aug 2009) | 1 line

fix legend bug ignoring CircleCollection
------------------------------------------------------------------------
r7402 | jdh2358 | 2009-08-06 10:10:45 -0500 (Thu, 06 Aug 2009) | 1 line

Mark Bakker <ma...@gm...> writes:

> I recently started using Eclipse with Pydev. I like it a lot but have not
> been able to get interactive plotting going (which otherwise works
> fine).

Does Pydev have any sort of specific support for matplotlib? I'm asking
because even in the normal Python shell interactive work can be
difficult, depending on the backend, and whatever command-loop magic
IPython does makes it work better. You could try different backends, but
I suspect that Pydev would have to have specific support for integrating
the backend-specific command loop for it to really work.

-- 
Jouni K. Seppänen
http://www.iki.fi/jks

Hello list,
I recently started using Eclipse with Pydev. I like it a lot but have not
been able to get interactive plotting going (which otherwise works fine).

My file is simple:

from pylab import *

ion()

plot([1,2,3])


When I run this form within Eclipse, I do see a graphing window open up but
then it disappears again when it is done, before I can even look at it.

Any suggestions on how to keep the graphing window open?

Thanks,

Mark

On Sun, Sep 20, 2009 at 11:45 PM, bwgoudey <bwg...@gm...> wrote:
>
> I'm trying to draw a scatter plot where each point is a different colour and
> is only a single pixel. Unfortunately at the minute, while I can change the
> colour and size, I haven't been able to remove the edges of the markers and
> my single pixel becomes the colour of the edge. I've attached some sample
> code below. Can anyway see what I've done wrong? It currently just gives
> back an error.
>

Your code works as expected with the current svn head. If you're using
older version of mpl, can you try to update and see if it solves the
problem?
If you're using recent version of mpl, please post an error message you got.

> And a related question. Is using scatter the best way to draw a plot of
> single pixels? Or would it be more efficient to use one of the more
> primitive classes to do this?
>

matplotlib basically creates a vector-based plot, and a notion of
"single-pixel" is not very meaningful. While it makes some sense for
raster output formats like png, but still, the result depends on the
dpi of your output. Note that unit for the parameter "s" in the
scatter function is in points(**2). 1 points corresponds to one pixel
in dpi=72. With higher dpi, the scatter points will occupy more than a
single pixel.

Regards,

-JJ


>
> fig = plt.figure()
> ax = fig.add_subplot(111)
>
> x=[1,2,3,4]
> y=[2,3,5,4]
> s=1
> c=[0.4, 0.2, 0.3, 0.5]
>
> prop=dict(edgecolors='none')
> ax.scatter(x,y, s=s, c=c, **prop)
> plt.show()
> --
> View this message in context: http://www.nabble.com/Changing-edge-colour-of-scatterplot-tp25530171p25530171.html
> Sent from the matplotlib - users mailing list archive at Nabble.com.
>
>
> ------------------------------------------------------------------------------
> Come build with us! The BlackBerry&reg; Developer Conference in SF, CA
> is the only developer event you need to attend this year. Jumpstart your
> developing skills, take BlackBerry mobile applications to market and stay
> ahead of the curve. Join us from November 9&#45;12, 2009. Register now&#33;
> http://p.sf.net/sfu/devconf
> _______________________________________________
> Matplotlib-users mailing list
> Mat...@li...
> https://lists.sourceforge.net/lists/listinfo/matplotlib-users
>