matplotlib-users Mailing List for matplotlib

On Sat, Oct 15, 2011 at 10:39 AM, <jos...@gm...> wrote:
> I'm building plots in stages using several different functions. Since
> the figure contains all information, I don't hand handles to
> individual elements around.
>
> What's the best way to check for a specific plot element? using
> isinstance, or are there specific attributes that could be checked?
Checkout out Artist.findobj
 http://matplotlib.sourceforge.net/api/artist_api.html#matplotlib.artist.Artist.findobj
Artist is the base class for everything in a matplotlib figure, the
Figure, Axes, Axis, Text, Line2D, Polygon, etc, all derive from
Artist, so you can call this method on any mpl object to recursively
search for objects of a given type. By recursive, I mean search the
objects children, their children, etc. See also
 http://matplotlib.sourceforge.net/users/artists.html
for more info on the artist hierarchy.
But for your specific example, each axes has an `images` attribute,
which is a list of images it contains (this is detailed in the artist
tutorial linked above, so you could simplify your code with something
like:
 images = np.concatenate([ax.images for ax in fig.axes])
JDH

I'm building plots in stages using several different functions. Since
the figure contains all information, I don't hand handles to
individual elements around.
What's the best way to check for a specific plot element? using
isinstance, or are there specific attributes that could be checked?
For example, I want to add a colorbar to the figure corresponding to
the first axis.imshow:
>>> images = [c for ax in f.axes for c in ax.get_children() if isinstance(c, mpl.image.AxesImage)]
>>> images
[<matplotlib.image.AxesImage object at 0x08C0CAD0>]
>>> f.colorbar(images[0])
<matplotlib.colorbar.Colorbar instance at 0x08E033F0>
>>> f
<matplotlib.figure.Figure object at 0x08B614D0>
example
using recipe http://matplotlib.sourceforge.net/examples/pylab_examples/subplots_adjust.html
https://lh5.googleusercontent.com/-rpw4NHbXvxM/TpmnumYNcRI/AAAAAAAAAK0/5zLnnUPjg0A/corrmatrixgrid.png
Thanks,
Josef

cool. it's done!! Thanks!!!
Chao
2011年10月15日 Tony Yu <ts...@gm...>
>
>
> On Sat, Oct 15, 2011 at 8:55 AM, Chao YUE <cha...@gm...> wrote:
>
>> Dear all,
>>
>> how can I make a scatter plot without edgecolor?
>>
>> import matplotlib.pyplot as plt
>> In [110]: plt.scatter(np.arange(10),np.arange(10,20),edgecolor=None)
>> Out[110]: <matplotlib.collections.CircleCollection object at 0x5cf16d0>
>>
>> in this case I can use edgecolor='w' to solve it, but when points overlap,
>> this does not work anymore.
>> Any help will be appreciated. Thanks.
>>
>
> Hi Chao,
>
> I think what you want is 'none':
>
> plt.scatter(np.arange(10),np.arange(10,20), color='y',edgecolor='none')
>
> It's confusing, but None is used to let matplotlib auto-select the color,
> while 'none' is used to turn off edge coloring.
>
> Best,
> -Tony
>
-- 
***********************************************************************************
Chao YUE
Laboratoire des Sciences du Climat et de l'Environnement (LSCE-IPSL)
UMR 1572 CEA-CNRS-UVSQ
Batiment 712 - Pe 119
91191 GIF Sur YVETTE Cedex
Tel: (33) 01 69 08 29 02; Fax:01.69.08.77.16
************************************************************************************

Figure.tight_layout() is a correct way.
Do you see that error only when you use Figure.tight_plot (and not
when you use plt.tight_layout)?
What happen you try the script below.
import matplotlib.pyplot as plt
fig = plt.figure(1)ax = fig.add_subplot(111)fig.tight_layout()
Regards,
-JJ
On Sat, Oct 15, 2011 at 9:13 AM, C M <cmp...@gm...> wrote:
> Just trying out the latest mpl 1.1.0 and the tight_layout() method. I saw
> the guide written about it, but am a unsure how to use this when using the
> OO approach to using Matplotlib.
>
> When using pyplot, the method is: plt.tight_layout(). When using the OO
> form of mpl, is it: figure.tight_layout() ?
>
> I assume it is, because I tried this and it didn't give me a name error, but
> I did get an error: ValueError: left cannot be >= right.
>
> What am I doing wrong?
>
> Thanks,
> Che
>
>
> ------------------------------------------------------------------------------
> All the data continuously generated in your IT infrastructure contains a
> definitive record of customers, application performance, security
> threats, fraudulent activity and more. Splunk takes this data and makes
> sense of it. Business sense. IT sense. Common sense.
> http://p.sf.net/sfu/splunk-d2d-oct
> _______________________________________________
> Matplotlib-users mailing list
> Mat...@li...
> https://lists.sourceforge.net/lists/listinfo/matplotlib-users
>
>

On Sat, Oct 15, 2011 at 8:55 AM, Chao YUE <cha...@gm...> wrote:
> Dear all,
>
> how can I make a scatter plot without edgecolor?
>
> import matplotlib.pyplot as plt
> In [110]: plt.scatter(np.arange(10),np.arange(10,20),edgecolor=None)
> Out[110]: <matplotlib.collections.CircleCollection object at 0x5cf16d0>
>
> in this case I can use edgecolor='w' to solve it, but when points overlap,
> this does not work anymore.
> Any help will be appreciated. Thanks.
>
Hi Chao,
I think what you want is 'none':
plt.scatter(np.arange(10),np.arange(10,20), color='y',edgecolor='none')
It's confusing, but None is used to let matplotlib auto-select the color,
while 'none' is used to turn off edge coloring.
Best,
-Tony

Hi Chao!
On Oct 15, 2011, at 2:55 PM, Chao YUE wrote:
> import matplotlib.pyplot as plt
> In [110]: plt.scatter(np.arange(10),np.arange(10,20),edgecolor=None)
> Out[110]: <matplotlib.collections.CircleCollection object at 0x5cf16d0>
> 
> in this case I can use edgecolor='w' to solve it, but when points overlap, this does not work anymore.
> Any help will be appreciated. Thanks.
For scatter plots, I usually use:
plt.plot(x,y,'.')
You can manipulate a bit the marker, markersize, style and alpha parameters (nice when there are really too many points overlapping, to get an idea of the "distribution"). In your ipython, type plot? to get all the available options.
Cheers,
jean-louis
> 
> Chao
> -- 
> ***********************************************************************************
> Chao YUE
> Laboratoire des Sciences du Climat et de l'Environnement (LSCE-IPSL)
> UMR 1572 CEA-CNRS-UVSQ
> Batiment 712 - Pe 119
> 91191 GIF Sur YVETTE Cedex
> Tel: (33) 01 69 08 29 02; Fax:01.69.08.77.16
> ************************************************************************************
> 
> ------------------------------------------------------------------------------
> All the data continuously generated in your IT infrastructure contains a
> definitive record of customers, application performance, security
> threats, fraudulent activity and more. Splunk takes this data and makes
> sense of it. Business sense. IT sense. Common sense.
> http://p.sf.net/sfu/splunk-d2d-oct_______________________________________________
> Matplotlib-users mailing list
> Mat...@li...
> https://lists.sourceforge.net/lists/listinfo/matplotlib-users

Dear all,
how can I make a scatter plot without edgecolor?
import matplotlib.pyplot as plt
In [110]: plt.scatter(np.arange(10),np.arange(10,20),edgecolor=None)
Out[110]: <matplotlib.collections.CircleCollection object at 0x5cf16d0>
in this case I can use edgecolor='w' to solve it, but when points overlap,
this does not work anymore.
Any help will be appreciated. Thanks.
Chao
-- 
***********************************************************************************
Chao YUE
Laboratoire des Sciences du Climat et de l'Environnement (LSCE-IPSL)
UMR 1572 CEA-CNRS-UVSQ
Batiment 712 - Pe 119
91191 GIF Sur YVETTE Cedex
Tel: (33) 01 69 08 29 02; Fax:01.69.08.77.16
************************************************************************************

On Friday, October 14, 2011, jopeto <g_n...@ho...> wrote:
>
> Hello,
>
> I have a question regarding setting a custom axis range. Here's a basic
> example:
>
> import matplotlib.pyplot as plt
>
> fig = plt.figure()
> ax = fig.add_subplot(1,1,1)
> ax.plot([1,2,3])
> ax.axis([xmin,xmax,ymin,ymax])
> plt.show()
>
> Now the question that I'm having is the following. The way I understand it
> now, I either have to live with the default axis ranges which matplotlib
> gives me, or I need to change all four values xmin, xmax, ymin, ymax using
> the ax.axis command. Say that I'm happy with xmin, ymin and ymax, and I
only
> want to set xmax to a value slightly larger than the default. Is there a
way
> to specify just that and to be able to keep the other three? The command
> which I'm looking for would be something like:
>
> ax.axis([keep default, xmax, keep default, keep default])
>
> If someone has a suggestion, I would be very grateful.
Use ax.set_xlim() and it's y-axis equivalent. You can even use kwargs to
those functions to manually set only one of the two limits.
Ben Root

Just trying out the latest mpl 1.1.0 and the tight_layout() method. I saw
the guide written about it, but am a unsure how to use this when using the
OO approach to using Matplotlib.
When using pyplot, the method is: plt.tight_layout(). When using the OO
form of mpl, is it: figure.tight_layout() ?
I assume it is, because I tried this and it didn't give me a name error, but
I did get an error: ValueError: left cannot be >= right.
What am I doing wrong?
Thanks,
Che