matplotlib-users Mailing List for matplotlib (Page 26)

No offense, but it really is outdated. Consider that it'll take two years to do the writing and the lecture work the research material is form 2007 to 2008. We now are in 2015. As you can tell from other books which have been published between 2013 and a really helpy book from March, 24th 2015 (yes, Benjamin Root wrote it), even they don't cover latest enhancements up to six month before print, (which might be seen a reasonable since changing is easy in a digitized world like ours). 
A good tutorial for the once, who do not have much experience in this field (I count myself in with the just one and a half year of experience in gui programming) is two things, actual up to six month to a year and straight forward, meaning It tells you what to do and doesn't bother you with design thoughts, API explanations nor tries to teach you programming. 
I have that book in my possession, but it didn't turn out to be helpful if you do not have the time do read it in whole. If you have the time to spin freely, you still will have conquered 80% by yourself and because it is still outdated for pyhton3 and matplotlib 1.4.3 the use is questionable.
cheers,Christian
 --
"A little learning never caused anyone's head to explode!"

"Ein wenig Lernen hat noch niemandens Kopf zum Explodieren gebracht!" 


     On Wednesday, April 15, 2015 3:44 AM, Chris O'Halloran <cm...@gm...> wrote:
   

 Can I recommend this book.  It was very helpful to me in figuring much of this out.

https://www.packtpub.com/application-development/matplotlib-python-developers

On 14 April 2015 at 18:14, Christian Ambros <am...@ym...> wrote:

Hi Ryan,
wow! This tutorial is one of the best I ever encountered. Nothing is missing, nothing is cryptic or unclear. What I like best is, that it get's along without using Qt Designer plugins or something similar strange. It's a good basis to start. Maybe you should write a book, covering all the untold things one needs to solve problems like that. I browsed through plenty of books the last weeks and what really is missing, is a cookbook about Qt Designer, Glade and wxWidgets and how to fill it with python3 and it's lib's like matplotlib, pyqtgraph, numpy, sympy etc.
I would buy it right away!cheers,Christian
 --
"A little learning never caused anyone's head to explode!"

"Ein wenig Lernen hat noch niemandens Kopf zum Explodieren gebracht!" 


     On Friday, April 10, 2015 7:14 PM, Ryan Nelson <rne...@gm...> wrote:
   

 Christian,
As it turns out, I wrote a blog post (for my terrible blog) about using Designer to create a MPL based GUI (http://blog.rcnelson.com/building-a-matplotlib-gui-with-qt-designer-part-1/). I was going to write this up for the MPL docs... But it got really long (3 parts), so I just used my personal site. It got so long because this was the second time I needed to figure this out, and I wanted to make a very detailed outline for my own future reference. Unfortunately, I don't have any experience with Qt5, but I imagine things are similar. I think they just rearranged the locations of some of the widgets, but I'd be curious to hear your experience. I gave up on PyQtdesignerplugins. I think it makes more sense to just use a generic widget as the MPL container. 
I would be very happy if you had comments for my Qt designer posts.
Ryan
On Wed, Apr 8, 2015 at 5:47 AM, Christian Ambros <am...@ym...> wrote:

Hi Ryan,
could you write down, as a tutorial, how you built the example with the qt designer?In the last hours I read all most everything what can be found on the issue of getting matplotlib running with pyqt5 and the designer but as you realized yourself, there is little to be found handy.
I'm stuck at a project, which has to use python3, and pyqt5 and am not allowed by my boss to fall back to pyqt4 or qt_compat. He wants to make sure that we use the latest revisions.
So I#m very pleased to read that someone already set food on this terrain. 
Qt5.4.1 is running and I installed PyQtdesingerplugins, in mind that they were written for PyQt4. Are they usable in 5? I added the env-variables to my bashrc, did get any changes shown in the designer. Of course I did a re-log-in to start fresh, but any changes were noteable.What possible ways of embedding matplotlib into a designer base pyqt5-gui else, are there?
cheers,Christian


 --
"A little learning never caused anyone's head to explode!"

"Ein wenig Lernen hat noch niemandens Kopf zum Explodieren gebracht!" 


     On Wednesday, February 18, 2015 11:59 PM, Ryan Nelson <rne...@gm...> wrote:
   

 Hello list,
A couple months ago, I spent quite a bit of time trying to figure out how to use Qt designer create a GUI with an embedded MPL window. Unfortunately, the Scipy cookbook page (http://wiki.scipy.org/Cookbook/Matplotlib/Qt_with_IPython_and_Designer) is very outdated. A recent post (http://matplotlib.1069221.n5.nabble.com/Re-Keep-list-of-figures-or-plots-and-flip-through-list-using-UI-td44961.html) brought up some questions about a use case very similar to mine, so I redid my example and was going to write a quick tutorial for the docs.
Unfortunately, I'm not a Qt guru, so I thought that I would ask on the list for some advice.  The OP and I were both interested in being able to have a list of figures that you could select from to change the plot window. The embedding examples in the docs create subclasses of FigureClass* and embed the plotting figure/axes/etc. This works but gets tricky, though, when trying to switch plots. Also, for interactive IPython work, I didn't like that the plotting objects were mixed in with all the QtGui.QWidget attributes, which makes introspective searching painful. My solution was to create a dictionary of matplotlib.figure.Figure objects that had all of the plotting stuff defined. Then when I select a new plot from the list, the old one is removed and a new FigureClass object is created using the selected Figure object. Has anyone else successfully done something like this? Is there a better way? Also, it seems if I zoom the current plot, change to a new plot, and change back, the zoom region is retained. Anyone know how to reset the zoom region?
Attached is my example: "window.py" is the Designer-created main window and "custommpl.py" is the subclass of the main window that I wrote. It's about as short as I could make it.

Thanks
Ryan


------------------------------------------------------------------------------
Download BIRT iHub F-Type - The Free Enterprise-Grade BIRT Server
from Actuate! Instantly Supercharge Your Business Reports and Dashboards
with Interactivity, Sharing, Native Excel Exports, App Integration & more
Get technology previously reserved for billion-dollar corporations, FREE
http://pubads.g.doubleclick.net/gampad/clk?id=190641631&iu=/4140/ostg.clktrk
_______________________________________________
Matplotlib-users mailing list
Mat...@li...
https://lists.sourceforge.net/lists/listinfo/matplotlib-users


   



   
------------------------------------------------------------------------------
BPM Camp - Free Virtual Workshop May 6th at 10am PDT/1PM EDT
Develop your own process in accordance with the BPMN 2 standard
Learn Process modeling best practices with Bonita BPM through live exercises
http://www.bonitasoft.com/be-part-of-it/events/bpm-camp-virtual- event?utm_
source=Sourceforge_BPM_Camp_5_6_15&utm_medium=email&utm_campaign=VA_SF
_______________________________________________
Matplotlib-users mailing list
Mat...@li...
https://lists.sourceforge.net/lists/listinfo/matplotlib-users




------------------------------------------------------------------------------
BPM Camp - Free Virtual Workshop May 6th at 10am PDT/1PM EDT
Develop your own process in accordance with the BPMN 2 standard
Learn Process modeling best practices with Bonita BPM through live exercises
http://www.bonitasoft.com/be-part-of-it/events/bpm-camp-virtual- event?utm_
source=Sourceforge_BPM_Camp_5_6_15&utm_medium=email&utm_campaign=VA_SF
_______________________________________________
Matplotlib-users mailing list
Mat...@li...
https://lists.sourceforge.net/lists/listinfo/matplotlib-users


  

Can I recommend this book.  It was very helpful to me in figuring much of
this out.

https://www.packtpub.com/application-development/matplotlib-python-developers

On 14 April 2015 at 18:14, Christian Ambros <am...@ym...> wrote:

> Hi Ryan,
>
> wow! This tutorial is one of the best I ever encountered. Nothing is
> missing, nothing is cryptic or unclear. What I like best is, that it get's
> along without using Qt Designer plugins or something similar strange. It's
> a good basis to start. Maybe you should write a book, covering all the
> untold things one needs to solve problems like that. I browsed through
> plenty of books the last weeks and what really is missing, is a cookbook
> about Qt Designer, Glade and wxWidgets and how to fill it with python3 and
> it's lib's like matplotlib, pyqtgraph, numpy, sympy etc.
>
> I would buy it right away!
> cheers,
> Christian
>
> --
> "A little learning never caused anyone's head to explode!"
>
>
> "Ein wenig Lernen hat noch niemandens Kopf zum Explodieren gebracht!"
>
>
>
>   On Friday, April 10, 2015 7:14 PM, Ryan Nelson <rne...@gm...>
> wrote:
>
>
> Christian,
>
> As it turns out, I wrote a blog post (for my terrible blog) about using
> Designer to create a MPL based GUI (
> http://blog.rcnelson.com/building-a-matplotlib-gui-with-qt-designer-part-1/).
> I was going to write this up for the MPL docs... But it got really long (3
> parts), so I just used my personal site. It got so long because this was
> the second time I needed to figure this out, and I wanted to make a very
> detailed outline for my own future reference. Unfortunately, I don't have
> any experience with Qt5, but I imagine things are similar. I think they
> just rearranged the locations of some of the widgets, but I'd be curious to
> hear your experience. I gave up on PyQtdesignerplugins. I think it makes
> more sense to just use a generic widget as the MPL container.
>
> I would be very happy if you had comments for my Qt designer posts.
>
> Ryan
>
> On Wed, Apr 8, 2015 at 5:47 AM, Christian Ambros <am...@ym...>
> wrote:
>
> Hi Ryan,
>
> could you write down, as a tutorial, how you built the example with the qt
> designer?
> In the last hours I read all most everything what can be found on the
> issue of getting matplotlib running with pyqt5 and the designer but as you
> realized yourself, there is little to be found handy.
>
> I'm stuck at a project, which has to use python3, and pyqt5 and am not
> allowed by my boss to fall back to pyqt4 or qt_compat. He wants to make
> sure that we use the latest revisions.
>
> So I#m very pleased to read that someone already set food on this terrain.
> Qt5.4.1 is running and I installed PyQtdesingerplugins, in mind that they
> were written for PyQt4. Are they usable in 5? I added the env-variables to
> my bashrc, did get any changes shown in the designer. Of course I did a
> re-log-in to start fresh, but any changes were noteable.
> What possible ways of embedding matplotlib into a designer base pyqt5-gui
> else, are there?
>
> cheers,
> Christian
>
>
>
> --
> "A little learning never caused anyone's head to explode!"
>
>
> "Ein wenig Lernen hat noch niemandens Kopf zum Explodieren gebracht!"
>
>
>
>   On Wednesday, February 18, 2015 11:59 PM, Ryan Nelson <
> rne...@gm...> wrote:
>
>
> Hello list,
>
> A couple months ago, I spent quite a bit of time trying to figure out how
> to use Qt designer create a GUI with an embedded MPL window. Unfortunately,
> the Scipy cookbook page (
> http://wiki.scipy.org/Cookbook/Matplotlib/Qt_with_IPython_and_Designer)
> is very outdated. A recent post (
> http://matplotlib.1069221.n5.nabble.com/Re-Keep-list-of-figures-or-plots-and-flip-through-list-using-UI-td44961.html)
> brought up some questions about a use case very similar to mine, so I redid
> my example and was going to write a quick tutorial for the docs.
>
> Unfortunately, I'm not a Qt guru, so I thought that I would ask on the
> list for some advice.  The OP and I were both interested in being able to
> have a list of figures that you could select from to change the plot
> window. The embedding examples in the docs create subclasses of
> FigureClass* and embed the plotting figure/axes/etc. This works but gets
> tricky, though, when trying to switch plots. Also, for interactive IPython
> work, I didn't like that the plotting objects were mixed in with all the
> QtGui.QWidget attributes, which makes introspective searching painful. My
> solution was to create a dictionary of matplotlib.figure.Figure objects
> that had all of the plotting stuff defined. Then when I select a new plot
> from the list, the old one is removed and a new FigureClass object is
> created using the selected Figure object. Has anyone else successfully done
> something like this? Is there a better way? Also, it seems if I zoom the
> current plot, change to a new plot, and change back, the zoom region is
> retained. Anyone know how to reset the zoom region?
>
> Attached is my example: "window.py" is the Designer-created main window
> and "custommpl.py" is the subclass of the main window that I wrote. It's
> about as short as I could make it.
>
> Thanks
>
> Ryan
>
>
>
>
> ------------------------------------------------------------------------------
> Download BIRT iHub F-Type - The Free Enterprise-Grade BIRT Server
> from Actuate! Instantly Supercharge Your Business Reports and Dashboards
> with Interactivity, Sharing, Native Excel Exports, App Integration & more
> Get technology previously reserved for billion-dollar corporations, FREE
>
> http://pubads.g.doubleclick.net/gampad/clk?id=190641631&iu=/4140/ostg.clktrk
> _______________________________________________
> Matplotlib-users mailing list
> Mat...@li...
> https://lists.sourceforge.net/lists/listinfo/matplotlib-users
>
>
>
>
>
>
>
> ------------------------------------------------------------------------------
> BPM Camp - Free Virtual Workshop May 6th at 10am PDT/1PM EDT
> Develop your own process in accordance with the BPMN 2 standard
> Learn Process modeling best practices with Bonita BPM through live
> exercises
> http://www.bonitasoft.com/be-part-of-it/events/bpm-camp-virtual-
> event?utm_
> source=Sourceforge_BPM_Camp_5_6_15&utm_medium=email&utm_campaign=VA_SF
> _______________________________________________
> Matplotlib-users mailing list
> Mat...@li...
> https://lists.sourceforge.net/lists/listinfo/matplotlib-users
>
>

Hi,

Here's another fractal music video, made with matplotlib:

https://www.youtube.com/watch?v=ANftKlKDtXc

Enjoy!

--Prahas

Le mardi 14 avril 2015 à 18:15 +0800, oyster a écrit :
> I am using anaconda(Python 2.7.6 |Anaconda 1.9.2 (32-bit)| (default,
> Nov 11 2013, 10:50:31) [MSC v.1500 32 bit (Intel)] on win32) on
> windows 7 64 bits
> And the matplotlib is 1.4.3, numpy is 1.9.2, and scipy is 0.15.1,
> which are all been updated by 'conda update xx'
> 
> As http://matplotlib.org/api/pyplot_api.html says
> [quote]
> matplotlib.pyplot.imread(*args, **kwargs) Read an image from a file
> into an array.
> 
> Return value is a numpy.array. For grayscale images, the return array
> is MxN. For RGB images, the return value is MxNx3. For RGBA images the
> return value is MxNx4.
> [/quote]
> 
> But if I read a 128*128*8 BPP gray PNG file, the array.shape is (128,
> 128, 3); if I read a 128*128*24BPP color PNG file, the array.shape is
> (128, 128, 4)

Are you sure your png "24bpp" does not have a transparency channel?

-- 
Fabrice

I am using anaconda(Python 2.7.6 |Anaconda 1.9.2 (32-bit)| (default,
Nov 11 2013, 10:50:31) [MSC v.1500 32 bit (Intel)] on win32) on
windows 7 64 bits
And the matplotlib is 1.4.3, numpy is 1.9.2, and scipy is 0.15.1,
which are all been updated by 'conda update xx'

As http://matplotlib.org/api/pyplot_api.html says
[quote]
matplotlib.pyplot.imread(*args, **kwargs) Read an image from a file
into an array.

Return value is a numpy.array. For grayscale images, the return array
is MxN. For RGB images, the return value is MxNx3. For RGBA images the
return value is MxNx4.
[/quote]

But if I read a 128*128*8 BPP gray PNG file, the array.shape is (128,
128, 3); if I read a 128*128*24BPP color PNG file, the array.shape is
(128, 128, 4)

Why? Thanks

[code]
from pylab import *

imgGrayPng=imread('python-gray.png')
print (imgGrayPng.shape)    #(128, 128, 3)


imgGrayJpg=imread('python-gray.jpg')
print (imgGrayJpg.shape)    #(128, 128)


imgColorPng=imread('python-color.png')
print (imgColorPng.shape)  #(128, 128, 4)


imgColorJpg=imread('python-color.jpg')
print (imgColorJpg.shape) #(128, 128, 3)


[/code]

Hi Ryan,
wow! This tutorial is one of the best I ever encountered. Nothing is missing, nothing is cryptic or unclear. What I like best is, that it get's along without using Qt Designer plugins or something similar strange. It's a good basis to start. Maybe you should write a book, covering all the untold things one needs to solve problems like that. I browsed through plenty of books the last weeks and what really is missing, is a cookbook about Qt Designer, Glade and wxWidgets and how to fill it with python3 and it's lib's like matplotlib, pyqtgraph, numpy, sympy etc.
I would buy it right away!cheers,Christian
 --
"A little learning never caused anyone's head to explode!"

"Ein wenig Lernen hat noch niemandens Kopf zum Explodieren gebracht!" 


     On Friday, April 10, 2015 7:14 PM, Ryan Nelson <rne...@gm...> wrote:
   

 Christian,
As it turns out, I wrote a blog post (for my terrible blog) about using Designer to create a MPL based GUI (http://blog.rcnelson.com/building-a-matplotlib-gui-with-qt-designer-part-1/). I was going to write this up for the MPL docs... But it got really long (3 parts), so I just used my personal site. It got so long because this was the second time I needed to figure this out, and I wanted to make a very detailed outline for my own future reference. Unfortunately, I don't have any experience with Qt5, but I imagine things are similar. I think they just rearranged the locations of some of the widgets, but I'd be curious to hear your experience. I gave up on PyQtdesignerplugins. I think it makes more sense to just use a generic widget as the MPL container. 
I would be very happy if you had comments for my Qt designer posts.
Ryan
On Wed, Apr 8, 2015 at 5:47 AM, Christian Ambros <am...@ym...> wrote:

Hi Ryan,
could you write down, as a tutorial, how you built the example with the qt designer?In the last hours I read all most everything what can be found on the issue of getting matplotlib running with pyqt5 and the designer but as you realized yourself, there is little to be found handy.
I'm stuck at a project, which has to use python3, and pyqt5 and am not allowed by my boss to fall back to pyqt4 or qt_compat. He wants to make sure that we use the latest revisions.
So I#m very pleased to read that someone already set food on this terrain. 
Qt5.4.1 is running and I installed PyQtdesingerplugins, in mind that they were written for PyQt4. Are they usable in 5? I added the env-variables to my bashrc, did get any changes shown in the designer. Of course I did a re-log-in to start fresh, but any changes were noteable.What possible ways of embedding matplotlib into a designer base pyqt5-gui else, are there?
cheers,Christian


 --
"A little learning never caused anyone's head to explode!"

"Ein wenig Lernen hat noch niemandens Kopf zum Explodieren gebracht!" 


     On Wednesday, February 18, 2015 11:59 PM, Ryan Nelson <rne...@gm...> wrote:
   

 Hello list,
A couple months ago, I spent quite a bit of time trying to figure out how to use Qt designer create a GUI with an embedded MPL window. Unfortunately, the Scipy cookbook page (http://wiki.scipy.org/Cookbook/Matplotlib/Qt_with_IPython_and_Designer) is very outdated. A recent post (http://matplotlib.1069221.n5.nabble.com/Re-Keep-list-of-figures-or-plots-and-flip-through-list-using-UI-td44961.html) brought up some questions about a use case very similar to mine, so I redid my example and was going to write a quick tutorial for the docs.
Unfortunately, I'm not a Qt guru, so I thought that I would ask on the list for some advice.  The OP and I were both interested in being able to have a list of figures that you could select from to change the plot window. The embedding examples in the docs create subclasses of FigureClass* and embed the plotting figure/axes/etc. This works but gets tricky, though, when trying to switch plots. Also, for interactive IPython work, I didn't like that the plotting objects were mixed in with all the QtGui.QWidget attributes, which makes introspective searching painful. My solution was to create a dictionary of matplotlib.figure.Figure objects that had all of the plotting stuff defined. Then when I select a new plot from the list, the old one is removed and a new FigureClass object is created using the selected Figure object. Has anyone else successfully done something like this? Is there a better way? Also, it seems if I zoom the current plot, change to a new plot, and change back, the zoom region is retained. Anyone know how to reset the zoom region?
Attached is my example: "window.py" is the Designer-created main window and "custommpl.py" is the subclass of the main window that I wrote. It's about as short as I could make it.

Thanks
Ryan


------------------------------------------------------------------------------
Download BIRT iHub F-Type - The Free Enterprise-Grade BIRT Server
from Actuate! Instantly Supercharge Your Business Reports and Dashboards
with Interactivity, Sharing, Native Excel Exports, App Integration & more
Get technology previously reserved for billion-dollar corporations, FREE
http://pubads.g.doubleclick.net/gampad/clk?id=190641631&iu=/4140/ostg.clktrk
_______________________________________________
Matplotlib-users mailing list
Mat...@li...
https://lists.sourceforge.net/lists/listinfo/matplotlib-users


   



  

animation objects have a private _stop() method. That might have to be a
workaround.

On Sun, Apr 12, 2015 at 9:24 AM, Thomas Caswell <tca...@gm...> wrote:

> You can
>
>
> ```
>
> #import matplotlib
>
> #matplotlib.use('nbagg')
>
> #%matplotlib nbagg
>
> import numpy as np
>
> import matplotlib.pyplot as plt
>
> import matplotlib.animation as animate
>
>
> class Testing(object):
>
>     def __init__(self, ):
>
>         self.fig = plt.figure()
>
>         array = np.random.rand(4,5)
>
>         array = np.zeros((4,5))
>
>         self.pc = plt.pcolor(array, edgecolor='k', linewidth=1.,
> animated=True)
>
>         self.pc.set_clim([0, 1])
>
>         self.points = [plt.scatter(np.random.rand(), np.random.rand(),
> animated=True)]
>
>
>     def update(self, iter_num):
>
>         array = np.random.rand(4*5)
>
>         self.pc.set_array(array)
>
>         for point in self.points:
>
>             point.set_offsets([np.random.rand(), np.random.rand()])
>
>
>         return (self.pc, ) + tuple(self.points)
>
>
>
> test = Testing()
>
> ani = animate.FuncAnimation(test.fig, test.update, interval=10,
> blit=False, frames=50)
>
> plt.show()
>
> ```
>
> note the addition of the `set_clim` line in the `__init__` method.
>
>
> You can also update the scatter artist in-place.  The other changes will
> make it a bit for performant if you use bliting (which does not work with
> nbagg currently)
>
> Sorry I missed that part of the question first time through.
>
> Tom
>
> On Sun, Apr 12, 2015, 08:31 Ryan Nelson <rne...@gm...> wrote:
>
>> Tom,
>>
>> Thanks for the links. It does seem like fragments of my problem are
>> addressed in each of those comments, so I guess I'll have to wait for a bit
>> until those things get resolved. For now, I can just tell my students to
>> restart the IPython kernel each time they run the animation, which isn't
>> that hard. It's too bad that there isn't a 'stop' method now, but it's good
>> to hear that it isn't a completely terrible idea.
>>
>> I do still need help with Question #3 from my original email, though,
>> because it affects both the Qt and nbagg backends, and it is a bit of a
>> show stopper. I can't quite understand why initializing a pcolor(mesh) with
>> random numbers makes it possible to update the array in an animation, but
>> if you use all zeros or ones, it seems to be immutable.
>>
>> Ryan
>>
>> On Sat, Apr 11, 2015 at 8:35 PM, Thomas Caswell <tca...@gm...>
>> wrote:
>>
>>> Ryan,
>>>
>>> I have not looked at your exact issue yet, but there seems to be some
>>> underlying issues with animation and nbagg which we have not tracked down
>>> yet. See:
>>>
>>> https://github.com/matplotlib/matplotlib/pull/4290
>>> https://github.com/matplotlib/matplotlib/issues/4287
>>> https://github.com/matplotlib/matplotlib/issues/4288
>>>
>>> Running until a given condition is an interesting idea, but I think that
>>> means the animation objects needs to have a public 'stop' method first!
>>>
>>> Tom
>>>
>>> On Fri, Apr 10, 2015 at 3:00 PM Ryan Nelson <rne...@gm...>
>>> wrote:
>>>
>>>> Good afternoon, all!
>>>>
>>>> I'm really digging the nbagg backend, and I'm trying to use it to make
>>>> an animation. As the subject suggests, though, I'm having some issues with
>>>> these features. I'm using Python 3.4, Matplotlib 1.4.3, and IPython 3.1.
>>>> Below is a small code sample that emulates my system. The pcolor call can
>>>> be substituted for pcolormesh, and I see the same behavior. (Sorry this is
>>>> a bit long. I tried to break it up as best as possible.)
>>>>
>>>> #############
>>>> #import matplotlib
>>>> #matplotlib.use('nbagg')
>>>> #%matplotlib nbagg
>>>> import numpy as np
>>>> import matplotlib.pyplot as plt
>>>> import matplotlib.animation as animate
>>>>
>>>> class Testing(object):
>>>>     def __init__(self, ):
>>>>         self.fig = plt.figure()
>>>>         array = np.random.rand(4,5)
>>>>         #array = np.zeros((4,5))
>>>>         self.pc = plt.pcolor(array, edgecolor='k', linewidth=1.)
>>>>         self.points = [plt.scatter(np.random.rand(), np.random.rand())]
>>>>
>>>>     def update(self, iter_num):
>>>>         array = np.random.rand(4*5)
>>>>         self.pc.set_array(array)
>>>>         for point in self.points:
>>>>             point.remove()
>>>>         self.points = [plt.scatter(np.random.rand(), np.random.rand())]
>>>>
>>>> test = Testing()
>>>> animate.FuncAnimation(test.fig, test.update, interval=1000, blit=False)
>>>> plt.show()
>>>> ###############
>>>>
>>>> 1. As is, this code runs fine with a Qt backend. It also runs fine as a
>>>> first call in a notebook if the `show` call is commented out and the
>>>> `%matplotlib` line is uncommented. However, if the `show` call is left in
>>>> and the `matplotlib.use` call is uncommented, then the pcolor array
>>>> changes, but the scatterpoint only shows on the first update and then
>>>> disappears forever. What is the difference between these two invocations?
>>>>
>>>> 2. With the `%matplotlib` magic uncommented and `show` removed, the
>>>> first invocation of this as a cell works fine. Closing the figure (with the
>>>> red X) and running the cell again shows two scatter plot points. Running it
>>>> a third time shows three scatter plot points. If you call `plt.clf` in the
>>>> next cell, I get a series of errors as follows:
>>>> _____
>>>> ERROR:tornado.application:Exception in callback <bound method
>>>> TimerTornado._on_timer of <matplotlib.backends.backend_nbagg.TimerTornado
>>>> object at 0x7f894cb10f98>>
>>>> Traceback (most recent call last):
>>>>   File "/usr/lib64/python3.4/site-packages/tornado/ioloop.py", line
>>>> 976, in _run
>>>>     return self.callback()
>>>>   File
>>>> "/usr/lib64/python3.4/site-packages/matplotlib/backend_bases.py", line
>>>> 1290, in _on_timer
>>>>     ret = func(*args, **kwargs)
>>>>   File "/usr/lib64/python3.4/site-packages/matplotlib/animation.py",
>>>> line 925, in _step
>>>>     still_going = Animation._step(self, *args)
>>>>   File "/usr/lib64/python3.4/site-packages/matplotlib/animation.py",
>>>> line 784, in _step
>>>>     self._draw_next_frame(framedata, self._blit)
>>>>   File "/usr/lib64/python3.4/site-packages/matplotlib/animation.py",
>>>> line 803, in _draw_next_frame
>>>>     self._draw_frame(framedata)
>>>>   File "/usr/lib64/python3.4/site-packages/matplotlib/animation.py",
>>>> line 1106, in _draw_frame
>>>>     self._drawn_artists = self._func(framedata, *self._args)
>>>>   File "<ipython-input-2-f9290d8f6154>", line 22, in update
>>>>     point.remove()
>>>>   File "/usr/lib64/python3.4/site-packages/matplotlib/artist.py", line
>>>> 139, in remove
>>>>     self._remove_method(self)
>>>>   File "/usr/lib64/python3.4/site-packages/matplotlib/axes/_base.py",
>>>> line 1479, in <lambda>
>>>>     collection._remove_method = lambda h: self.collections.remove(h)
>>>> ValueError: list.remove(x): x not in list
>>>> ______
>>>> Why does this happen? Is there a way to close the animation cleanly?
>>>>
>>>> 3. If I uncomment the `np.zeros` call, the pcolor array never updates
>>>> irrespective of the backend. I see the same behavior with `np.ones` as
>>>> well, even if the dtype is set to `float`. Is there are a way to start with
>>>> a all-zero pcolor that allow dynamic updates?
>>>>
>>>> 4. I'd like to be able to have the animation run until a certain
>>>> condition is met. Is there a way to code a clean break for the animation?
>>>>
>>>>
>>>> As always, any help is most appreciated!
>>>>
>>>> Ryan
>>>>
>>>>
>>>>
>>>>
>>>>
>>>> ------------------------------------------------------------
>>>> ------------------
>>>> BPM Camp - Free Virtual Workshop May 6th at 10am PDT/1PM EDT
>>>> Develop your own process in accordance with the BPMN 2 standard
>>>> Learn Process modeling best practices with Bonita BPM through live
>>>> exercises
>>>> http://www.bonitasoft.com/be-part-of-it/events/bpm-camp-virtual-
>>>> event?utm_
>>>> source=Sourceforge_BPM_Camp_5_6_15&utm_medium=email&utm_
>>>> campaign=VA_SF_______________________________________________
>>>> Matplotlib-users mailing list
>>>> Mat...@li...
>>>> https://lists.sourceforge.net/lists/listinfo/matplotlib-users
>>>>
>>>
>>
>
> ------------------------------------------------------------------------------
> BPM Camp - Free Virtual Workshop May 6th at 10am PDT/1PM EDT
> Develop your own process in accordance with the BPMN 2 standard
> Learn Process modeling best practices with Bonita BPM through live
> exercises
> http://www.bonitasoft.com/be-part-of-it/events/bpm-camp-virtual-
> event?utm_
> source=Sourceforge_BPM_Camp_5_6_15&utm_medium=email&utm_campaign=VA_SF
> _______________________________________________
> Matplotlib-users mailing list
> Mat...@li...
> https://lists.sourceforge.net/lists/listinfo/matplotlib-users
>
>

Thanks for the report, I turned it into a github issue: 
https://github.com/matplotlib/matplotlib/issues/4331

On 2015/04/05 11:19 PM, giacomo boffi wrote:
> INTRO
> =====
>
> please consider the following code (I'm trying to draw a timeline)
>
> 1 from matplotlib import pyplot, patches
> 2 fig = pyplot.figure()
> 3 ax = fig.add_subplot('111')
> 4 ax.add_patch(patches.Rectangle((1933,0.25), 73, 0.5))
> 5 pyplot.show()
>
> that gives me a plot with the x axis that goes from 0.0 to 1.0,
> now consider
>
>    ...
> 5 ax.set_xlim((1933,1933+73))
> 6 pyplot.show()
>
> this gives me an x axis that goes _exactly_ from 1933 to 2006,
> eventually drawing a line superposed to the lower spine
>
>    ...
> 5 ax.plot((1933,1933+73),(0,0))
> 6 pyplot.show()
>
> gives me what I really want, that is an x axis running from 1930 to
> 2010, with the limits automatically rounded by matplotlib...
>
> (I noted that the extra line forces a rounding also for the y axis
> limits, but that's not a problem...)
>
> QUESTION
> ========
>
> I want matplotlib to round the limits of the x axis automatically,
> when given explicitly the lower and upper limits of the data, how to?

I think the initial problem is that ax.add_patch() is not triggering the 
autoscaling that you are looking for; the higher-level plot() function 
does so.  After your call to ax.add_patch(), try adding 
ax.autoscale_view().

Eric


>
> Thank you in advance
>

You can


```

#import matplotlib

#matplotlib.use('nbagg')

#%matplotlib nbagg

import numpy as np

import matplotlib.pyplot as plt

import matplotlib.animation as animate


class Testing(object):

    def __init__(self, ):

        self.fig = plt.figure()

        array = np.random.rand(4,5)

        array = np.zeros((4,5))

        self.pc = plt.pcolor(array, edgecolor='k', linewidth=1.,
animated=True)

        self.pc.set_clim([0, 1])

        self.points = [plt.scatter(np.random.rand(), np.random.rand(),
animated=True)]


    def update(self, iter_num):

        array = np.random.rand(4*5)

        self.pc.set_array(array)

        for point in self.points:

            point.set_offsets([np.random.rand(), np.random.rand()])


        return (self.pc, ) + tuple(self.points)



test = Testing()

ani = animate.FuncAnimation(test.fig, test.update, interval=10, blit=False,
frames=50)

plt.show()

```

note the addition of the `set_clim` line in the `__init__` method.


You can also update the scatter artist in-place.  The other changes will
make it a bit for performant if you use bliting (which does not work with
nbagg currently)

Sorry I missed that part of the question first time through.

Tom

On Sun, Apr 12, 2015, 08:31 Ryan Nelson <rne...@gm...> wrote:

> Tom,
>
> Thanks for the links. It does seem like fragments of my problem are
> addressed in each of those comments, so I guess I'll have to wait for a bit
> until those things get resolved. For now, I can just tell my students to
> restart the IPython kernel each time they run the animation, which isn't
> that hard. It's too bad that there isn't a 'stop' method now, but it's good
> to hear that it isn't a completely terrible idea.
>
> I do still need help with Question #3 from my original email, though,
> because it affects both the Qt and nbagg backends, and it is a bit of a
> show stopper. I can't quite understand why initializing a pcolor(mesh) with
> random numbers makes it possible to update the array in an animation, but
> if you use all zeros or ones, it seems to be immutable.
>
> Ryan
>
> On Sat, Apr 11, 2015 at 8:35 PM, Thomas Caswell <tca...@gm...>
> wrote:
>
>> Ryan,
>>
>> I have not looked at your exact issue yet, but there seems to be some
>> underlying issues with animation and nbagg which we have not tracked down
>> yet. See:
>>
>> https://github.com/matplotlib/matplotlib/pull/4290
>> https://github.com/matplotlib/matplotlib/issues/4287
>> https://github.com/matplotlib/matplotlib/issues/4288
>>
>> Running until a given condition is an interesting idea, but I think that
>> means the animation objects needs to have a public 'stop' method first!
>>
>> Tom
>>
>> On Fri, Apr 10, 2015 at 3:00 PM Ryan Nelson <rne...@gm...>
>> wrote:
>>
>>> Good afternoon, all!
>>>
>>> I'm really digging the nbagg backend, and I'm trying to use it to make
>>> an animation. As the subject suggests, though, I'm having some issues with
>>> these features. I'm using Python 3.4, Matplotlib 1.4.3, and IPython 3.1.
>>> Below is a small code sample that emulates my system. The pcolor call can
>>> be substituted for pcolormesh, and I see the same behavior. (Sorry this is
>>> a bit long. I tried to break it up as best as possible.)
>>>
>>> #############
>>> #import matplotlib
>>> #matplotlib.use('nbagg')
>>> #%matplotlib nbagg
>>> import numpy as np
>>> import matplotlib.pyplot as plt
>>> import matplotlib.animation as animate
>>>
>>> class Testing(object):
>>>     def __init__(self, ):
>>>         self.fig = plt.figure()
>>>         array = np.random.rand(4,5)
>>>         #array = np.zeros((4,5))
>>>         self.pc = plt.pcolor(array, edgecolor='k', linewidth=1.)
>>>         self.points = [plt.scatter(np.random.rand(), np.random.rand())]
>>>
>>>     def update(self, iter_num):
>>>         array = np.random.rand(4*5)
>>>         self.pc.set_array(array)
>>>         for point in self.points:
>>>             point.remove()
>>>         self.points = [plt.scatter(np.random.rand(), np.random.rand())]
>>>
>>> test = Testing()
>>> animate.FuncAnimation(test.fig, test.update, interval=1000, blit=False)
>>> plt.show()
>>> ###############
>>>
>>> 1. As is, this code runs fine with a Qt backend. It also runs fine as a
>>> first call in a notebook if the `show` call is commented out and the
>>> `%matplotlib` line is uncommented. However, if the `show` call is left in
>>> and the `matplotlib.use` call is uncommented, then the pcolor array
>>> changes, but the scatterpoint only shows on the first update and then
>>> disappears forever. What is the difference between these two invocations?
>>>
>>> 2. With the `%matplotlib` magic uncommented and `show` removed, the
>>> first invocation of this as a cell works fine. Closing the figure (with the
>>> red X) and running the cell again shows two scatter plot points. Running it
>>> a third time shows three scatter plot points. If you call `plt.clf` in the
>>> next cell, I get a series of errors as follows:
>>> _____
>>> ERROR:tornado.application:Exception in callback <bound method
>>> TimerTornado._on_timer of <matplotlib.backends.backend_nbagg.TimerTornado
>>> object at 0x7f894cb10f98>>
>>> Traceback (most recent call last):
>>>   File "/usr/lib64/python3.4/site-packages/tornado/ioloop.py", line 976,
>>> in _run
>>>     return self.callback()
>>>   File "/usr/lib64/python3.4/site-packages/matplotlib/backend_bases.py",
>>> line 1290, in _on_timer
>>>     ret = func(*args, **kwargs)
>>>   File "/usr/lib64/python3.4/site-packages/matplotlib/animation.py",
>>> line 925, in _step
>>>     still_going = Animation._step(self, *args)
>>>   File "/usr/lib64/python3.4/site-packages/matplotlib/animation.py",
>>> line 784, in _step
>>>     self._draw_next_frame(framedata, self._blit)
>>>   File "/usr/lib64/python3.4/site-packages/matplotlib/animation.py",
>>> line 803, in _draw_next_frame
>>>     self._draw_frame(framedata)
>>>   File "/usr/lib64/python3.4/site-packages/matplotlib/animation.py",
>>> line 1106, in _draw_frame
>>>     self._drawn_artists = self._func(framedata, *self._args)
>>>   File "<ipython-input-2-f9290d8f6154>", line 22, in update
>>>     point.remove()
>>>   File "/usr/lib64/python3.4/site-packages/matplotlib/artist.py", line
>>> 139, in remove
>>>     self._remove_method(self)
>>>   File "/usr/lib64/python3.4/site-packages/matplotlib/axes/_base.py",
>>> line 1479, in <lambda>
>>>     collection._remove_method = lambda h: self.collections.remove(h)
>>> ValueError: list.remove(x): x not in list
>>> ______
>>> Why does this happen? Is there a way to close the animation cleanly?
>>>
>>> 3. If I uncomment the `np.zeros` call, the pcolor array never updates
>>> irrespective of the backend. I see the same behavior with `np.ones` as
>>> well, even if the dtype is set to `float`. Is there are a way to start with
>>> a all-zero pcolor that allow dynamic updates?
>>>
>>> 4. I'd like to be able to have the animation run until a certain
>>> condition is met. Is there a way to code a clean break for the animation?
>>>
>>>
>>> As always, any help is most appreciated!
>>>
>>> Ryan
>>>
>>>
>>>
>>>
>>>
>>> ------------------------------------------------------------
>>> ------------------
>>> BPM Camp - Free Virtual Workshop May 6th at 10am PDT/1PM EDT
>>> Develop your own process in accordance with the BPMN 2 standard
>>> Learn Process modeling best practices with Bonita BPM through live
>>> exercises
>>> http://www.bonitasoft.com/be-part-of-it/events/bpm-camp-virtual-
>>> event?utm_
>>> source=Sourceforge_BPM_Camp_5_6_15&utm_medium=email&utm_
>>> campaign=VA_SF_______________________________________________
>>> Matplotlib-users mailing list
>>> Mat...@li...
>>> https://lists.sourceforge.net/lists/listinfo/matplotlib-users
>>>
>>
>

Tom,

Thanks for the links. It does seem like fragments of my problem are
addressed in each of those comments, so I guess I'll have to wait for a bit
until those things get resolved. For now, I can just tell my students to
restart the IPython kernel each time they run the animation, which isn't
that hard. It's too bad that there isn't a 'stop' method now, but it's good
to hear that it isn't a completely terrible idea.

I do still need help with Question #3 from my original email, though,
because it affects both the Qt and nbagg backends, and it is a bit of a
show stopper. I can't quite understand why initializing a pcolor(mesh) with
random numbers makes it possible to update the array in an animation, but
if you use all zeros or ones, it seems to be immutable.

Ryan

On Sat, Apr 11, 2015 at 8:35 PM, Thomas Caswell <tca...@gm...> wrote:

> Ryan,
>
> I have not looked at your exact issue yet, but there seems to be some
> underlying issues with animation and nbagg which we have not tracked down
> yet. See:
>
> https://github.com/matplotlib/matplotlib/pull/4290
> https://github.com/matplotlib/matplotlib/issues/4287
> https://github.com/matplotlib/matplotlib/issues/4288
>
> Running until a given condition is an interesting idea, but I think that
> means the animation objects needs to have a public 'stop' method first!
>
> Tom
>
> On Fri, Apr 10, 2015 at 3:00 PM Ryan Nelson <rne...@gm...> wrote:
>
>> Good afternoon, all!
>>
>> I'm really digging the nbagg backend, and I'm trying to use it to make an
>> animation. As the subject suggests, though, I'm having some issues with
>> these features. I'm using Python 3.4, Matplotlib 1.4.3, and IPython 3.1.
>> Below is a small code sample that emulates my system. The pcolor call can
>> be substituted for pcolormesh, and I see the same behavior. (Sorry this is
>> a bit long. I tried to break it up as best as possible.)
>>
>> #############
>> #import matplotlib
>> #matplotlib.use('nbagg')
>> #%matplotlib nbagg
>> import numpy as np
>> import matplotlib.pyplot as plt
>> import matplotlib.animation as animate
>>
>> class Testing(object):
>>     def __init__(self, ):
>>         self.fig = plt.figure()
>>         array = np.random.rand(4,5)
>>         #array = np.zeros((4,5))
>>         self.pc = plt.pcolor(array, edgecolor='k', linewidth=1.)
>>         self.points = [plt.scatter(np.random.rand(), np.random.rand())]
>>
>>     def update(self, iter_num):
>>         array = np.random.rand(4*5)
>>         self.pc.set_array(array)
>>         for point in self.points:
>>             point.remove()
>>         self.points = [plt.scatter(np.random.rand(), np.random.rand())]
>>
>> test = Testing()
>> animate.FuncAnimation(test.fig, test.update, interval=1000, blit=False)
>> plt.show()
>> ###############
>>
>> 1. As is, this code runs fine with a Qt backend. It also runs fine as a
>> first call in a notebook if the `show` call is commented out and the
>> `%matplotlib` line is uncommented. However, if the `show` call is left in
>> and the `matplotlib.use` call is uncommented, then the pcolor array
>> changes, but the scatterpoint only shows on the first update and then
>> disappears forever. What is the difference between these two invocations?
>>
>> 2. With the `%matplotlib` magic uncommented and `show` removed, the first
>> invocation of this as a cell works fine. Closing the figure (with the red
>> X) and running the cell again shows two scatter plot points. Running it a
>> third time shows three scatter plot points. If you call `plt.clf` in the
>> next cell, I get a series of errors as follows:
>> _____
>> ERROR:tornado.application:Exception in callback <bound method
>> TimerTornado._on_timer of <matplotlib.backends.backend_nbagg.TimerTornado
>> object at 0x7f894cb10f98>>
>> Traceback (most recent call last):
>>   File "/usr/lib64/python3.4/site-packages/tornado/ioloop.py", line 976,
>> in _run
>>     return self.callback()
>>   File "/usr/lib64/python3.4/site-packages/matplotlib/backend_bases.py",
>> line 1290, in _on_timer
>>     ret = func(*args, **kwargs)
>>   File "/usr/lib64/python3.4/site-packages/matplotlib/animation.py", line
>> 925, in _step
>>     still_going = Animation._step(self, *args)
>>   File "/usr/lib64/python3.4/site-packages/matplotlib/animation.py", line
>> 784, in _step
>>     self._draw_next_frame(framedata, self._blit)
>>   File "/usr/lib64/python3.4/site-packages/matplotlib/animation.py", line
>> 803, in _draw_next_frame
>>     self._draw_frame(framedata)
>>   File "/usr/lib64/python3.4/site-packages/matplotlib/animation.py", line
>> 1106, in _draw_frame
>>     self._drawn_artists = self._func(framedata, *self._args)
>>   File "<ipython-input-2-f9290d8f6154>", line 22, in update
>>     point.remove()
>>   File "/usr/lib64/python3.4/site-packages/matplotlib/artist.py", line
>> 139, in remove
>>     self._remove_method(self)
>>   File "/usr/lib64/python3.4/site-packages/matplotlib/axes/_base.py",
>> line 1479, in <lambda>
>>     collection._remove_method = lambda h: self.collections.remove(h)
>> ValueError: list.remove(x): x not in list
>> ______
>> Why does this happen? Is there a way to close the animation cleanly?
>>
>> 3. If I uncomment the `np.zeros` call, the pcolor array never updates
>> irrespective of the backend. I see the same behavior with `np.ones` as
>> well, even if the dtype is set to `float`. Is there are a way to start with
>> a all-zero pcolor that allow dynamic updates?
>>
>> 4. I'd like to be able to have the animation run until a certain
>> condition is met. Is there a way to code a clean break for the animation?
>>
>>
>> As always, any help is most appreciated!
>>
>> Ryan
>>
>>
>>
>>
>>
>> ------------------------------------------------------------
>> ------------------
>> BPM Camp - Free Virtual Workshop May 6th at 10am PDT/1PM EDT
>> Develop your own process in accordance with the BPMN 2 standard
>> Learn Process modeling best practices with Bonita BPM through live
>> exercises
>> http://www.bonitasoft.com/be-part-of-it/events/bpm-camp-virtual-
>> event?utm_
>> source=Sourceforge_BPM_Camp_5_6_15&utm_medium=email&utm_
>> campaign=VA_SF_______________________________________________
>> Matplotlib-users mailing list
>> Mat...@li...
>> https://lists.sourceforge.net/lists/listinfo/matplotlib-users
>>
>

I am not really sure what you mean by 'round'.  Do you want to suppress the
offset or do you want mpl to pick 'nice' values after you have explicitly
set the limits?

Tom

On Mon, Apr 6, 2015 at 5:27 AM giacomo boffi <gia...@gm...>
wrote:

> INTRO
> =====
>
> please consider the following code (I'm trying to draw a timeline)
>
> 1 from matplotlib import pyplot, patches
> 2 fig = pyplot.figure()
> 3 ax = fig.add_subplot('111')
> 4 ax.add_patch(patches.Rectangle((1933,0.25), 73, 0.5))
> 5 pyplot.show()
>
> that gives me a plot with the x axis that goes from 0.0 to 1.0,
> now consider
>
>   ...
> 5 ax.set_xlim((1933,1933+73))
> 6 pyplot.show()
>
> this gives me an x axis that goes _exactly_ from 1933 to 2006,
> eventually drawing a line superposed to the lower spine
>
>   ...
> 5 ax.plot((1933,1933+73),(0,0))
> 6 pyplot.show()
>
> gives me what I really want, that is an x axis running from 1930 to
> 2010, with the limits automatically rounded by matplotlib...
>
> (I noted that the extra line forces a rounding also for the y axis
> limits, but that's not a problem...)
>
> QUESTION
> ========
>
> I want matplotlib to round the limits of the x axis automatically,
> when given explicitly the lower and upper limits of the data, how to?
>
> Thank you in advance
>
> --
> "We have met the enemy and he is us."
>                                                         --- Pogo.
>
>
> ------------------------------------------------------------
> ------------------
> BPM Camp - Free Virtual Workshop May 6th at 10am PDT/1PM EDT
> Develop your own process in accordance with the BPMN 2 standard
> Learn Process modeling best practices with Bonita BPM through live
> exercises
> http://www.bonitasoft.com/be-part-of-it/events/bpm-camp-virtual-
> event?utm_
> source=Sourceforge_BPM_Camp_5_6_15&utm_medium=email&utm_campaign=VA_SF
> _______________________________________________
> Matplotlib-users mailing list
> Mat...@li...
> https://lists.sourceforge.net/lists/listinfo/matplotlib-users
>

Malk,

This is a bit of a gap in mpl currently (but has come up a couple of times (
https://github.com/matplotlib/matplotlib/issues/4217,
https://github.com/matplotlib/matplotlib/issues/2203, and
http://matplotlib.org/devdocs/devel/MEP/MEP24.html).

One of the hold ups has been lack of a developer that uses polar plots
day-to-day and a lack of really clear use cases.  I think we have three (at
least) distinct use cases.

 1. origin always an 0, negative radius rotates by pi, always full 2pi
around, always solid circle (no inner axes) (useful for plotting bunches of
vectors against each other)
 2. center is at arbitrary 'r', values less than 'origin' are just not
shown, always full 2pi, no inner axes (use for for dB plots showing power
as function of angle)
 3. inner axes with arbitrary origin, possibly not full 2pi

It is not immediately clear to me if these can all be done with the same
projection or even if they can be done with the 'standard' Axes class or if
we need to user AxesArtist here.

This discussion should probably move to MEP24/the devel list.

What version of mpl are you using?  Your example causes seg-faults (!) on
my system, but I have not sorted out why (may be really strange install
issues on my end).

Tom

On Thu, Apr 9, 2015 at 6:08 AM Maik Hoffmann <mai...@b-...> wrote:

> Hello,
> I'm using mpl_toolkits.axisartist.floating_axes.GridHelperCurveLinear
> for creating half-polar plots from 180 degree measurements for receive
> sensitivity.
>
> Working with the measurement values itself is no problem if I let the
> values scaling start at zero.
> If I use normalized values I can plot it also, but if I transform it
> into the dB scale I got a segfault in this lib.
>
> I provide an example. For my problems I would like to have a solution
> that I can either use r limit from -30 to 0 (f3) or changing the tick
> labels in figure f2.
>
> And by the way is there a possibility that the if i want to plot data in
> the range from 80 to 120, that rlim(80,120) would set the 80 to the
> centerpoint? At the moment I got only a small stripe.
>
> [code]
> """Demo of polar plot of arbitrary theta. This is a workaround for MPL's
> polar plot limitation
> to a full 360 deg.
>
> Based on
> http://matplotlib.org/mpl_toolkits/axes_grid/examples/
> demo_floating_axes.py
>
> get from
> https://github.com/neuropy/neuropy/blob/master/neuropy/
> scripts/polar_demo.py
> TODO: license / copyright
> """
>
> from __future__ import division
> from __future__ import print_function
>
> import numpy as np
> import matplotlib.pyplot as plt
>
> from matplotlib.transforms import Affine2D
> from matplotlib.projections import PolarAxes
> from mpl_toolkits.axisartist import angle_helper
> from mpl_toolkits.axisartist.grid_finder import MaxNLocator
> from mpl_toolkits.axisartist.floating_axes import GridHelperCurveLinear,
> FloatingSubplot
>
>
> def fractional_polar_axes(f, thlim=(0, 180), rlim=(0, 1), step=(30, 0.2),
>                            thlabel='theta', rlabel='r', ticklabels=True,
> theta_offset=0):
>      """Return polar axes that adhere to desired theta (in deg) and r
> limits. steps for theta
>      and r are really just hints for the locators."""
>      th0, th1 = thlim # deg
>      r0, r1 = rlim
>      thstep, rstep = step
>
>      tr_rotate = Affine2D().translate(theta_offset, 0)
>      # scale degrees to radians:
>      tr_scale = Affine2D().scale(np.pi/180., 1.)
>      #pa = axes(polar="true") # Create a polar axis
>      pa = PolarAxes
>      tr = tr_rotate + tr_scale + pa.PolarTransform()
>      theta_grid_locator = angle_helper.LocatorDMS((th1-th0)//thstep)
>      r_grid_locator = MaxNLocator((r1-r0)//rstep)
>      theta_tick_formatter = angle_helper.FormatterDMS()
>
>      grid_helper = GridHelperCurveLinear(tr,
>                                          extremes=(th0, th1, r0, r1),
>                                          grid_locator1=theta_grid_locator,
>                                          grid_locator2=r_grid_locator,
>
> tick_formatter1=theta_tick_formatter,
>                                          tick_formatter2=None)
>
>      a = FloatingSubplot(f, 111, grid_helper=grid_helper)
>
>      f.add_subplot(a)
>
>      # adjust x axis (theta):
>      a.axis["bottom"].set_visible(False)
>      a.axis["top"].set_axis_direction("bottom") # tick direction
>      a.axis["top"].toggle(ticklabels=ticklabels, label=bool(thlabel))
>      a.axis["top"].major_ticklabels.set_axis_direction("top")
>      a.axis["top"].label.set_axis_direction("top")
>
>      # adjust y axis (r):
>      a.axis["left"].set_axis_direction("bottom") # tick direction
>      a.axis["right"].set_axis_direction("top") # tick direction
>      a.axis["left"].toggle(ticklabels=ticklabels, label=bool(rlabel))
>
>      # add labels:
>      a.axis["top"].label.set_text(thlabel)
>      a.axis["left"].label.set_text(rlabel)
>
>      # create a parasite axes whose transData is theta, r:
>      auxa = a.get_aux_axes(tr)
>      # make aux_ax to have a clip path as in a?:
>      auxa.patch = a.patch
>      # this has a side effect that the patch is drawn twice, and
> possibly over some other
>      # artists. So, we decrease the zorder a bit to prevent this:
>      a.patch.zorder = -2
>
>
>      # add sector lines for both dimensions:
>      thticks = grid_helper.grid_info['lon_info'][0]
>      rticks = grid_helper.grid_info['lat_info'][0]
>      for th in thticks[1:-1]: # all but the first and last
>          auxa.plot([th, th], [r0, r1], '--', c='grey', zorder=-1)
>      for ri, r in enumerate(rticks):
>          # plot first r line as axes border in solid black only if it
> isn't at r=0
>          if ri == 0 and r != 0:
>              ls, lw, color = 'solid', 2, 'black'
>          else:
>              ls, lw, color = 'dashed', 1, 'grey'
>          # From http://stackoverflow.com/a/19828753/2020363
>          auxa.add_artist(plt.Circle([0, 0], radius=r, ls=ls, lw=lw,
> color=color, fill=False,
>                          transform=auxa.transData._b, zorder=-1))
>
>      return auxa
>
>
> if __name__ == '__main__':
>      f1 = plt.figure(facecolor='white', figsize=(16/2.54, 12/2.54),
> dpi=600)
>      a1 = fractional_polar_axes(f1, thlim=(-90, 90),step=(10,
> 0.2),theta_offset=90)
>      # example spiral plot:
>      thstep = 10
>      th = np.arange(-90, 90+thstep, thstep) # deg
>      rstep = 1/(len(th)-1)
>      r = np.arange(0, 1+rstep, rstep)
>      a1.plot(th, r, 'b')
>
>
>      f1.show()
>
>      f2 = plt.figure(facecolor='white', figsize=(16/2.54, 12/2.54),
> dpi=600)
>      a2 = fractional_polar_axes(f2, thlim=(-90,
> 90),rlim=(0,30),step=(10, 8),theta_offset=90)
>      # example spiral plot:
>      r2 = 20 * np.log10(r) +30
>      a2.plot(th, r2, 'b')
>
>      f2.show()
>
>      f3 = plt.figure(facecolor='white', figsize=(16/2.54, 12/2.54),
> dpi=600)
>      a3 = fractional_polar_axes(f2, thlim=(-90, 90),rlim=(-30,0),
> step=(10, 8),theta_offset=90)
>      # example spiral plot:
>      r3 = 20 * np.log10(r)
>      a3.plot(th, r2, 'b')
>
>      f2.show()
>
> [\code]
>
> --
>
>
> ------------------------------------------------------------
> ------------------
> BPM Camp - Free Virtual Workshop May 6th at 10am PDT/1PM EDT
> Develop your own process in accordance with the BPMN 2 standard
> Learn Process modeling best practices with Bonita BPM through live
> exercises
> http://www.bonitasoft.com/be-part-of-it/events/bpm-camp-virtual-
> event?utm_
> source=Sourceforge_BPM_Camp_5_6_15&utm_medium=email&utm_campaign=VA_SF
> _______________________________________________
> Matplotlib-users mailing list
> Mat...@li...
> https://lists.sourceforge.net/lists/listinfo/matplotlib-users
>

Ryan,

I have not looked at your exact issue yet, but there seems to be some
underlying issues with animation and nbagg which we have not tracked down
yet. See:

https://github.com/matplotlib/matplotlib/pull/4290
https://github.com/matplotlib/matplotlib/issues/4287
https://github.com/matplotlib/matplotlib/issues/4288

Running until a given condition is an interesting idea, but I think that
means the animation objects needs to have a public 'stop' method first!

Tom

On Fri, Apr 10, 2015 at 3:00 PM Ryan Nelson <rne...@gm...> wrote:

> Good afternoon, all!
>
> I'm really digging the nbagg backend, and I'm trying to use it to make an
> animation. As the subject suggests, though, I'm having some issues with
> these features. I'm using Python 3.4, Matplotlib 1.4.3, and IPython 3.1.
> Below is a small code sample that emulates my system. The pcolor call can
> be substituted for pcolormesh, and I see the same behavior. (Sorry this is
> a bit long. I tried to break it up as best as possible.)
>
> #############
> #import matplotlib
> #matplotlib.use('nbagg')
> #%matplotlib nbagg
> import numpy as np
> import matplotlib.pyplot as plt
> import matplotlib.animation as animate
>
> class Testing(object):
>     def __init__(self, ):
>         self.fig = plt.figure()
>         array = np.random.rand(4,5)
>         #array = np.zeros((4,5))
>         self.pc = plt.pcolor(array, edgecolor='k', linewidth=1.)
>         self.points = [plt.scatter(np.random.rand(), np.random.rand())]
>
>     def update(self, iter_num):
>         array = np.random.rand(4*5)
>         self.pc.set_array(array)
>         for point in self.points:
>             point.remove()
>         self.points = [plt.scatter(np.random.rand(), np.random.rand())]
>
> test = Testing()
> animate.FuncAnimation(test.fig, test.update, interval=1000, blit=False)
> plt.show()
> ###############
>
> 1. As is, this code runs fine with a Qt backend. It also runs fine as a
> first call in a notebook if the `show` call is commented out and the
> `%matplotlib` line is uncommented. However, if the `show` call is left in
> and the `matplotlib.use` call is uncommented, then the pcolor array
> changes, but the scatterpoint only shows on the first update and then
> disappears forever. What is the difference between these two invocations?
>
> 2. With the `%matplotlib` magic uncommented and `show` removed, the first
> invocation of this as a cell works fine. Closing the figure (with the red
> X) and running the cell again shows two scatter plot points. Running it a
> third time shows three scatter plot points. If you call `plt.clf` in the
> next cell, I get a series of errors as follows:
> _____
> ERROR:tornado.application:Exception in callback <bound method
> TimerTornado._on_timer of <matplotlib.backends.backend_nbagg.TimerTornado
> object at 0x7f894cb10f98>>
> Traceback (most recent call last):
>   File "/usr/lib64/python3.4/site-packages/tornado/ioloop.py", line 976,
> in _run
>     return self.callback()
>   File "/usr/lib64/python3.4/site-packages/matplotlib/backend_bases.py",
> line 1290, in _on_timer
>     ret = func(*args, **kwargs)
>   File "/usr/lib64/python3.4/site-packages/matplotlib/animation.py", line
> 925, in _step
>     still_going = Animation._step(self, *args)
>   File "/usr/lib64/python3.4/site-packages/matplotlib/animation.py", line
> 784, in _step
>     self._draw_next_frame(framedata, self._blit)
>   File "/usr/lib64/python3.4/site-packages/matplotlib/animation.py", line
> 803, in _draw_next_frame
>     self._draw_frame(framedata)
>   File "/usr/lib64/python3.4/site-packages/matplotlib/animation.py", line
> 1106, in _draw_frame
>     self._drawn_artists = self._func(framedata, *self._args)
>   File "<ipython-input-2-f9290d8f6154>", line 22, in update
>     point.remove()
>   File "/usr/lib64/python3.4/site-packages/matplotlib/artist.py", line
> 139, in remove
>     self._remove_method(self)
>   File "/usr/lib64/python3.4/site-packages/matplotlib/axes/_base.py", line
> 1479, in <lambda>
>     collection._remove_method = lambda h: self.collections.remove(h)
> ValueError: list.remove(x): x not in list
> ______
> Why does this happen? Is there a way to close the animation cleanly?
>
> 3. If I uncomment the `np.zeros` call, the pcolor array never updates
> irrespective of the backend. I see the same behavior with `np.ones` as
> well, even if the dtype is set to `float`. Is there are a way to start with
> a all-zero pcolor that allow dynamic updates?
>
> 4. I'd like to be able to have the animation run until a certain condition
> is met. Is there a way to code a clean break for the animation?
>
>
> As always, any help is most appreciated!
>
> Ryan
>
>
>
>
>
> ------------------------------------------------------------
> ------------------
> BPM Camp - Free Virtual Workshop May 6th at 10am PDT/1PM EDT
> Develop your own process in accordance with the BPMN 2 standard
> Learn Process modeling best practices with Bonita BPM through live
> exercises
> http://www.bonitasoft.com/be-part-of-it/events/bpm-camp-virtual-
> event?utm_
> source=Sourceforge_BPM_Camp_5_6_15&utm_medium=email&utm_
> campaign=VA_SF_______________________________________________
> Matplotlib-users mailing list
> Mat...@li...
> https://lists.sourceforge.net/lists/listinfo/matplotlib-users
>

We have a pr in to add ternary axes, but unfortunately I do not understand
it well enough to tell you of contours will work.

See https://github.com/matplotlib/matplotlib/pull/3828 and links with in.

Tom

On Sat, Apr 11, 2015, 07:47 nxkryptor nxkr <nxk...@gm...> wrote:

> I am trying to create ternary plots with matplotlib as shown in the figure
> (source
> <http://stackoverflow.com/questions/10879361/ternary-plot-and-filled-contour>).
> The axes are A, B, C and D values should be denoted by contours and the
> points need to be labelled like in figure.
>
> [image: enter image description here]Can such plots be created in
> matplotlib or with Python? This question is already posted on
> stackoverflow
> <http://stackoverflow.com/questions/29512046/how-to-create-ternary-contour-plot-in-python>
> .
>
>
> nxkr
>
> ------------------------------------------------------------------------------
> BPM Camp - Free Virtual Workshop May 6th at 10am PDT/1PM EDT
> Develop your own process in accordance with the BPMN 2 standard
> Learn Process modeling best practices with Bonita BPM through live
> exercises
> http://www.bonitasoft.com/be-part-of-it/events/bpm-camp-virtual-
> event?utm_
>
> source=Sourceforge_BPM_Camp_5_6_15&utm_medium=email&utm_campaign=VA_SF_______________________________________________
> Matplotlib-users mailing list
> Mat...@li...
> https://lists.sourceforge.net/lists/listinfo/matplotlib-users
>

I am trying to create ternary plots with matplotlib as shown in the figure (
source
<http://stackoverflow.com/questions/10879361/ternary-plot-and-filled-contour>).
The axes are A, B, C and D values should be denoted by contours and the
points need to be labelled like in figure.

[image: enter image description here]Can such plots be created in
matplotlib or with Python? This question is already posted on stackoverflow
<http://stackoverflow.com/questions/29512046/how-to-create-ternary-contour-plot-in-python>
.


nxkr

Christian,

As it turns out, I wrote a blog post (for my terrible blog) about using
Designer to create a MPL based GUI (
http://blog.rcnelson.com/building-a-matplotlib-gui-with-qt-designer-part-1/).
I was going to write this up for the MPL docs... But it got really long (3
parts), so I just used my personal site. It got so long because this was
the second time I needed to figure this out, and I wanted to make a very
detailed outline for my own future reference. Unfortunately, I don't have
any experience with Qt5, but I imagine things are similar. I think they
just rearranged the locations of some of the widgets, but I'd be curious to
hear your experience. I gave up on PyQtdesignerplugins. I think it makes
more sense to just use a generic widget as the MPL container.

I would be very happy if you had comments for my Qt designer posts.

Ryan

On Wed, Apr 8, 2015 at 5:47 AM, Christian Ambros <am...@ym...> wrote:

> Hi Ryan,
>
> could you write down, as a tutorial, how you built the example with the qt
> designer?
> In the last hours I read all most everything what can be found on the
> issue of getting matplotlib running with pyqt5 and the designer but as you
> realized yourself, there is little to be found handy.
>
> I'm stuck at a project, which has to use python3, and pyqt5 and am not
> allowed by my boss to fall back to pyqt4 or qt_compat. He wants to make
> sure that we use the latest revisions.
>
> So I#m very pleased to read that someone already set food on this terrain.
> Qt5.4.1 is running and I installed PyQtdesingerplugins, in mind that they
> were written for PyQt4. Are they usable in 5? I added the env-variables to
> my bashrc, did get any changes shown in the designer. Of course I did a
> re-log-in to start fresh, but any changes were noteable.
> What possible ways of embedding matplotlib into a designer base pyqt5-gui
> else, are there?
>
> cheers,
> Christian
>
>
>
> --
> "A little learning never caused anyone's head to explode!"
>
>
> "Ein wenig Lernen hat noch niemandens Kopf zum Explodieren gebracht!"
>
>
>
>   On Wednesday, February 18, 2015 11:59 PM, Ryan Nelson <
> rne...@gm...> wrote:
>
>
> Hello list,
>
> A couple months ago, I spent quite a bit of time trying to figure out how
> to use Qt designer create a GUI with an embedded MPL window. Unfortunately,
> the Scipy cookbook page (
> http://wiki.scipy.org/Cookbook/Matplotlib/Qt_with_IPython_and_Designer)
> is very outdated. A recent post (
> http://matplotlib.1069221.n5.nabble.com/Re-Keep-list-of-figures-or-plots-and-flip-through-list-using-UI-td44961.html)
> brought up some questions about a use case very similar to mine, so I redid
> my example and was going to write a quick tutorial for the docs.
>
> Unfortunately, I'm not a Qt guru, so I thought that I would ask on the
> list for some advice.  The OP and I were both interested in being able to
> have a list of figures that you could select from to change the plot
> window. The embedding examples in the docs create subclasses of
> FigureClass* and embed the plotting figure/axes/etc. This works but gets
> tricky, though, when trying to switch plots. Also, for interactive IPython
> work, I didn't like that the plotting objects were mixed in with all the
> QtGui.QWidget attributes, which makes introspective searching painful. My
> solution was to create a dictionary of matplotlib.figure.Figure objects
> that had all of the plotting stuff defined. Then when I select a new plot
> from the list, the old one is removed and a new FigureClass object is
> created using the selected Figure object. Has anyone else successfully done
> something like this? Is there a better way? Also, it seems if I zoom the
> current plot, change to a new plot, and change back, the zoom region is
> retained. Anyone know how to reset the zoom region?
>
> Attached is my example: "window.py" is the Designer-created main window
> and "custommpl.py" is the subclass of the main window that I wrote. It's
> about as short as I could make it.
>
> Thanks
>
> Ryan
>
>
>
>
> ------------------------------------------------------------------------------
> Download BIRT iHub F-Type - The Free Enterprise-Grade BIRT Server
> from Actuate! Instantly Supercharge Your Business Reports and Dashboards
> with Interactivity, Sharing, Native Excel Exports, App Integration & more
> Get technology previously reserved for billion-dollar corporations, FREE
>
> http://pubads.g.doubleclick.net/gampad/clk?id=190641631&iu=/4140/ostg.clktrk
> _______________________________________________
> Matplotlib-users mailing list
> Mat...@li...
> https://lists.sourceforge.net/lists/listinfo/matplotlib-users
>
>
>

Good afternoon, all!

I'm really digging the nbagg backend, and I'm trying to use it to make an
animation. As the subject suggests, though, I'm having some issues with
these features. I'm using Python 3.4, Matplotlib 1.4.3, and IPython 3.1.
Below is a small code sample that emulates my system. The pcolor call can
be substituted for pcolormesh, and I see the same behavior. (Sorry this is
a bit long. I tried to break it up as best as possible.)

#############
#import matplotlib
#matplotlib.use('nbagg')
#%matplotlib nbagg
import numpy as np
import matplotlib.pyplot as plt
import matplotlib.animation as animate

class Testing(object):
    def __init__(self, ):
        self.fig = plt.figure()
        array = np.random.rand(4,5)
        #array = np.zeros((4,5))
        self.pc = plt.pcolor(array, edgecolor='k', linewidth=1.)
        self.points = [plt.scatter(np.random.rand(), np.random.rand())]

    def update(self, iter_num):
        array = np.random.rand(4*5)
        self.pc.set_array(array)
        for point in self.points:
            point.remove()
        self.points = [plt.scatter(np.random.rand(), np.random.rand())]

test = Testing()
animate.FuncAnimation(test.fig, test.update, interval=1000, blit=False)
plt.show()
###############

1. As is, this code runs fine with a Qt backend. It also runs fine as a
first call in a notebook if the `show` call is commented out and the
`%matplotlib` line is uncommented. However, if the `show` call is left in
and the `matplotlib.use` call is uncommented, then the pcolor array
changes, but the scatterpoint only shows on the first update and then
disappears forever. What is the difference between these two invocations?

2. With the `%matplotlib` magic uncommented and `show` removed, the first
invocation of this as a cell works fine. Closing the figure (with the red
X) and running the cell again shows two scatter plot points. Running it a
third time shows three scatter plot points. If you call `plt.clf` in the
next cell, I get a series of errors as follows:
_____
ERROR:tornado.application:Exception in callback <bound method
TimerTornado._on_timer of <matplotlib.backends.backend_nbagg.TimerTornado
object at 0x7f894cb10f98>>
Traceback (most recent call last):
  File "/usr/lib64/python3.4/site-packages/tornado/ioloop.py", line 976, in
_run
    return self.callback()
  File "/usr/lib64/python3.4/site-packages/matplotlib/backend_bases.py",
line 1290, in _on_timer
    ret = func(*args, **kwargs)
  File "/usr/lib64/python3.4/site-packages/matplotlib/animation.py", line
925, in _step
    still_going = Animation._step(self, *args)
  File "/usr/lib64/python3.4/site-packages/matplotlib/animation.py", line
784, in _step
    self._draw_next_frame(framedata, self._blit)
  File "/usr/lib64/python3.4/site-packages/matplotlib/animation.py", line
803, in _draw_next_frame
    self._draw_frame(framedata)
  File "/usr/lib64/python3.4/site-packages/matplotlib/animation.py", line
1106, in _draw_frame
    self._drawn_artists = self._func(framedata, *self._args)
  File "<ipython-input-2-f9290d8f6154>", line 22, in update
    point.remove()
  File "/usr/lib64/python3.4/site-packages/matplotlib/artist.py", line 139,
in remove
    self._remove_method(self)
  File "/usr/lib64/python3.4/site-packages/matplotlib/axes/_base.py", line
1479, in <lambda>
    collection._remove_method = lambda h: self.collections.remove(h)
ValueError: list.remove(x): x not in list
______
Why does this happen? Is there a way to close the animation cleanly?

3. If I uncomment the `np.zeros` call, the pcolor array never updates
irrespective of the backend. I see the same behavior with `np.ones` as
well, even if the dtype is set to `float`. Is there are a way to start with
a all-zero pcolor that allow dynamic updates?

4. I'd like to be able to have the animation run until a certain condition
is met. Is there a way to code a clean break for the animation?


As always, any help is most appreciated!

Ryan

Hello,
I'm using mpl_toolkits.axisartist.floating_axes.GridHelperCurveLinear
for creating half-polar plots from 180 degree measurements for receive 
sensitivity.

Working with the measurement values itself is no problem if I let the 
values scaling start at zero.
If I use normalized values I can plot it also, but if I transform it 
into the dB scale I got a segfault in this lib.

I provide an example. For my problems I would like to have a solution 
that I can either use r limit from -30 to 0 (f3) or changing the tick 
labels in figure f2.

And by the way is there a possibility that the if i want to plot data in 
the range from 80 to 120, that rlim(80,120) would set the 80 to the 
centerpoint? At the moment I got only a small stripe.

[code]
"""Demo of polar plot of arbitrary theta. This is a workaround for MPL's 
polar plot limitation
to a full 360 deg.

Based on 
http://matplotlib.org/mpl_toolkits/axes_grid/examples/demo_floating_axes.py

get from 
https://github.com/neuropy/neuropy/blob/master/neuropy/scripts/polar_demo.py
TODO: license / copyright
"""

from __future__ import division
from __future__ import print_function

import numpy as np
import matplotlib.pyplot as plt

from matplotlib.transforms import Affine2D
from matplotlib.projections import PolarAxes
from mpl_toolkits.axisartist import angle_helper
from mpl_toolkits.axisartist.grid_finder import MaxNLocator
from mpl_toolkits.axisartist.floating_axes import GridHelperCurveLinear, 
FloatingSubplot


def fractional_polar_axes(f, thlim=(0, 180), rlim=(0, 1), step=(30, 0.2),
                           thlabel='theta', rlabel='r', ticklabels=True, 
theta_offset=0):
     """Return polar axes that adhere to desired theta (in deg) and r 
limits. steps for theta
     and r are really just hints for the locators."""
     th0, th1 = thlim # deg
     r0, r1 = rlim
     thstep, rstep = step

     tr_rotate = Affine2D().translate(theta_offset, 0)
     # scale degrees to radians:
     tr_scale = Affine2D().scale(np.pi/180., 1.)
     #pa = axes(polar="true") # Create a polar axis
     pa = PolarAxes
     tr = tr_rotate + tr_scale + pa.PolarTransform()
     theta_grid_locator = angle_helper.LocatorDMS((th1-th0)//thstep)
     r_grid_locator = MaxNLocator((r1-r0)//rstep)
     theta_tick_formatter = angle_helper.FormatterDMS()

     grid_helper = GridHelperCurveLinear(tr,
                                         extremes=(th0, th1, r0, r1),
                                         grid_locator1=theta_grid_locator,
                                         grid_locator2=r_grid_locator,
 
tick_formatter1=theta_tick_formatter,
                                         tick_formatter2=None)

     a = FloatingSubplot(f, 111, grid_helper=grid_helper)

     f.add_subplot(a)

     # adjust x axis (theta):
     a.axis["bottom"].set_visible(False)
     a.axis["top"].set_axis_direction("bottom") # tick direction
     a.axis["top"].toggle(ticklabels=ticklabels, label=bool(thlabel))
     a.axis["top"].major_ticklabels.set_axis_direction("top")
     a.axis["top"].label.set_axis_direction("top")

     # adjust y axis (r):
     a.axis["left"].set_axis_direction("bottom") # tick direction
     a.axis["right"].set_axis_direction("top") # tick direction
     a.axis["left"].toggle(ticklabels=ticklabels, label=bool(rlabel))

     # add labels:
     a.axis["top"].label.set_text(thlabel)
     a.axis["left"].label.set_text(rlabel)

     # create a parasite axes whose transData is theta, r:
     auxa = a.get_aux_axes(tr)
     # make aux_ax to have a clip path as in a?:
     auxa.patch = a.patch
     # this has a side effect that the patch is drawn twice, and 
possibly over some other
     # artists. So, we decrease the zorder a bit to prevent this:
     a.patch.zorder = -2


     # add sector lines for both dimensions:
     thticks = grid_helper.grid_info['lon_info'][0]
     rticks = grid_helper.grid_info['lat_info'][0]
     for th in thticks[1:-1]: # all but the first and last
         auxa.plot([th, th], [r0, r1], '--', c='grey', zorder=-1)
     for ri, r in enumerate(rticks):
         # plot first r line as axes border in solid black only if it 
isn't at r=0
         if ri == 0 and r != 0:
             ls, lw, color = 'solid', 2, 'black'
         else:
             ls, lw, color = 'dashed', 1, 'grey'
         # From http://stackoverflow.com/a/19828753/2020363
         auxa.add_artist(plt.Circle([0, 0], radius=r, ls=ls, lw=lw, 
color=color, fill=False,
                         transform=auxa.transData._b, zorder=-1))

     return auxa


if __name__ == '__main__':
     f1 = plt.figure(facecolor='white', figsize=(16/2.54, 12/2.54), dpi=600)
     a1 = fractional_polar_axes(f1, thlim=(-90, 90),step=(10, 
0.2),theta_offset=90)
     # example spiral plot:
     thstep = 10
     th = np.arange(-90, 90+thstep, thstep) # deg
     rstep = 1/(len(th)-1)
     r = np.arange(0, 1+rstep, rstep)
     a1.plot(th, r, 'b')


     f1.show()

     f2 = plt.figure(facecolor='white', figsize=(16/2.54, 12/2.54), dpi=600)
     a2 = fractional_polar_axes(f2, thlim=(-90, 
90),rlim=(0,30),step=(10, 8),theta_offset=90)
     # example spiral plot:
     r2 = 20 * np.log10(r) +30
     a2.plot(th, r2, 'b')

     f2.show()

     f3 = plt.figure(facecolor='white', figsize=(16/2.54, 12/2.54), dpi=600)
     a3 = fractional_polar_axes(f2, thlim=(-90, 90),rlim=(-30,0), 
step=(10, 8),theta_offset=90)
     # example spiral plot:
     r3 = 20 * np.log10(r)
     a3.plot(th, r2, 'b')

     f2.show()

[\code]

-- 

On 2015/04/08 11:15 AM, Jody Klymak wrote:
> Hi Eric,
>
>>> import matplotlib.pyplot as plt
>>> fig, axes = plt.subplots(nrows=2,sharex=True)
>>> axes[0].set_aspect(1.)
>>> axes[0].plot(np.arange(10),np.arange(10))
>>> axes[0].set_ylim([0,24])
>>> axes[0].set_xlim([0,12])
>>> axes[1].plot(np.arange(10),np.arange(10)*2.)
>>> plt.show()
>>>
>>> does not work as I'd expect.  axes[0]'s ylim gets changed so that the
>>> line is no longer viewable (= 10-14).  In my opinion, the two calls
>>> should work the same, except in the second case, axes[1]'s xlim
>>> should be 0-12.
>>>
>>
>> If you leave out the set_ylim call, it works.  Given that you have
>> specified set_ylim[0, 24], how is mpl supposed to know what ylim range
>> you really want, when the axis constraint means it can only plot a
>> small part of the specified range?
>
> It doesn't for me:
>
> import matplotlib.pyplot as plt
> fig, axes = plt.subplots(nrows=2,sharex=True)
> axes[0].set_aspect(1.)
> axes[0].plot(np.arange(10),np.arange(10))
> axes[1].plot(np.arange(10),np.arange(10)*2.)
> plt.show()
>
> still curtails the y limit in axes[0], in my case from ~2.9 to ~6.1 (see
> attached).

Jody, you told it you want x to go from 0 to 12, and have an aspect 
ratio of 1.  It is doing exactly that.  Are you expecting it to override 
your xlim specification?

>
>> Basically, you have set up conflicting constraints, and mpl failed
>> to resolve the conflict the way you think it should have.  Maybe that
>> could be improved, but I warn you, it's a tricky business.  Squeeze in
>> one place and things pop out somewhere else.
>
> I'm not clear what the conflicting constraints are.  There seems to be
> an unspoken one that sharex=True means the physical size of the axes
> must be the same. That constraint doesn't exist if sharex=False, and
> set_aspect() works as expected.  I'm questioning the unspoken
> constraint, and questioning why set_aspect() (or looking at the code
> apply_aspect()) needs to know about shared axes at all.  No doubt there
> is a use case I'm missing...

Yes, sharex=True means the x-axis is identical.  It's shared.  That's 
just what it means, and what it has always meant.  That's what it is 
for--to lock together the x axes of two or more Axes.

>
> I *can* see the issue if we think setting aspect ratios should *not*
> change the size of the axes, because changing the aspect ratio changes
> the data limits and then you have a problem checking all the shared axes
> to see which one has the largest data limits.  Its particularly
> problematic because I think that apply_aspect() is only called at
> draw()-time.  That seems a hard problem, but I'm not sure what the use
> case is for it, so I have trouble wrapping my head around it.

The use case for what--calling apply_aspect at draw time?  That's the 
only time it knows everything it needs to know in the general case, when 
there might be pan/zoom/reshape events.

Eric

>
> Thanks,   Jody
>
> --
> Jody Klymak
> http://web.uvic.ca/~jklymak/
>
>
>
>
>
>
>
> ------------------------------------------------------------------------------
> BPM Camp - Free Virtual Workshop May 6th at 10am PDT/1PM EDT
> Develop your own process in accordance with the BPMN 2 standard
> Learn Process modeling best practices with Bonita BPM through live exercises
> http://www.bonitasoft.com/be-part-of-it/events/bpm-camp-virtual- event?utm_
> source=Sourceforge_BPM_Camp_5_6_15&utm_medium=email&utm_campaign=VA_SF
>
>
>
> _______________________________________________
> Matplotlib-users mailing list
> Mat...@li...
> https://lists.sourceforge.net/lists/listinfo/matplotlib-users
>

Hi Eric,

>> import matplotlib.pyplot as plt
>> fig, axes = plt.subplots(nrows=2,sharex=True)
>> axes[0].set_aspect(1.)
>> axes[0].plot(np.arange(10),np.arange(10))
>> axes[0].set_ylim([0,24])
>> axes[0].set_xlim([0,12])
>> axes[1].plot(np.arange(10),np.arange(10)*2.)
>> plt.show()
>> 
>> does not work as I'd expect.  axes[0]'s ylim gets changed so that the line is no longer viewable (= 10-14).  In my opinion, the two calls should work the same, except in the second case, axes[1]'s xlim should be 0-12.
>> 
> 
> If you leave out the set_ylim call, it works.  Given that you have specified set_ylim[0, 24], how is mpl supposed to know what ylim range you really want, when the axis constraint means it can only plot a small part of the specified range?  

It doesn't for me:

import matplotlib.pyplot as plt
fig, axes = plt.subplots(nrows=2,sharex=True)
axes[0].set_aspect(1.)
axes[0].plot(np.arange(10),np.arange(10))
axes[1].plot(np.arange(10),np.arange(10)*2.)
plt.show()

still curtails the y limit in axes[0], in my case from ~2.9 to ~6.1 (see attached).

> Basically, you have set up conflicting constraints, and mpl failed to resolve the conflict the way you think it should have.  Maybe that could be improved, but I warn you, it's a tricky business.  Squeeze in one place and things pop out somewhere else.

I'm not clear what the conflicting constraints are.  There seems to be an unspoken one that sharex=True means the physical size of the axes must be the same. That constraint doesn't exist if sharex=False, and set_aspect() works as expected.  I'm questioning the unspoken constraint, and questioning why set_aspect() (or looking at the code apply_aspect()) needs to know about shared axes at all.  No doubt there is a use case I'm missing...

I *can* see the issue if we think setting aspect ratios should *not* change the size of the axes, because changing the aspect ratio changes the data limits and then you have a problem checking all the shared axes to see which one has the largest data limits.  Its particularly problematic because I think that apply_aspect() is only called at draw()-time.  That seems a hard problem, but I'm not sure what the use case is for it, so I have trouble wrapping my head around it. 

Thanks,   Jody


--
Jody Klymak    
http://web.uvic.ca/~jklymak/

On 2015/04/08 8:43 AM, Jody Klymak wrote:
> Hi Eric,
>
>> On 8 Apr 2015, at  11:02 AM, Eric Firing <ef...@ha...> wrote:
>>
>> I'm the guilty party for most of how set_aspect works.  I developed it a
>> long time ago. Yes, there was a reason--still is, I'm 99% sure--but I
>> don't remember everything, and don't want to take the time now to
>> reconstruct the whole rationale.  When I was developing the behavior, I
>> was paying a lot of attention to what happens under various scenarios of
>> resizing and reshaping the window, and turning options on and off.
>> There are some basic conflicts that arise when shared axes are combined
>> with fixed aspect ratios, autoscaling, and gui-driven reshaping.
>> Sometimes 'box-forced' does what people want, maybe more often than not;
>> but I'm pretty sure it can lead to trouble, which is the reason it is
>> not the default.
>
> OK, first my apologies:
>
> import matplotlib.pyplot as plt
> fig, axes = plt.subplots(nrows=2)
> axes[0].set_aspect(1.)
> axes[0].plot(np.arange(10),np.arange(10))
> axes[0].set_ylim([0,24])
> axes[0].set_xlim([0,12])
> axes[1].plot(np.arange(10),np.arange(10)*2.)
> plt.show()
>
> Works as I'd expect.  axes[0] gets shrunk in the x dimension to make the aspect ratio 1.
>
> However:
>
> import matplotlib.pyplot as plt
> fig, axes = plt.subplots(nrows=2,sharex=True)
> axes[0].set_aspect(1.)
> axes[0].plot(np.arange(10),np.arange(10))
> axes[0].set_ylim([0,24])
> axes[0].set_xlim([0,12])
> axes[1].plot(np.arange(10),np.arange(10)*2.)
> plt.show()
>
> does not work as I'd expect.  axes[0]'s ylim gets changed so that the line is no longer viewable (= 10-14).  In my opinion, the two calls should work the same, except in the second case, axes[1]'s xlim should be 0-12.
>

If you leave out the set_ylim call, it works.  Given that you have 
specified set_ylim[0, 24], how is mpl supposed to know what ylim range 
you really want, when the axis constraint means it can only plot a small 
part of the specified range?  Basically, you have set up conflicting 
constraints, and mpl failed to resolve the conflict the way you think it 
should have.  Maybe that could be improved, but I warn you, it's a 
tricky business.  Squeeze in one place and things pop out somewhere else.

Eric

> Even worse, if I use the same aspect ratio in axes[1], they *both* do not show all the data:
>
> axes[0].set_aspect(1.)
>
> It appears here that with sharex=True the shape of the axis no longer becomes mutable when set_aspect() is used, whereas if sharex=False set_aspect() can change the axis shape. I don't see any reason for sharex to foist this behaviour onto set_aspect().
>
> I guess the workaround is don't use sharex=True, but I actually think this is a bug.
>
> Thanks,   Jody
>
> --
> Jody Klymak
> http://web.uvic.ca/~jklymak/
>
>
>
>
>

Hi Eric,

> On 8 Apr 2015, at  11:02 AM, Eric Firing <ef...@ha...> wrote:
> 
> I'm the guilty party for most of how set_aspect works.  I developed it a 
> long time ago. Yes, there was a reason--still is, I'm 99% sure--but I 
> don't remember everything, and don't want to take the time now to 
> reconstruct the whole rationale.  When I was developing the behavior, I 
> was paying a lot of attention to what happens under various scenarios of 
> resizing and reshaping the window, and turning options on and off. 
> There are some basic conflicts that arise when shared axes are combined 
> with fixed aspect ratios, autoscaling, and gui-driven reshaping. 
> Sometimes 'box-forced' does what people want, maybe more often than not; 
> but I'm pretty sure it can lead to trouble, which is the reason it is 
> not the default.

OK, first my apologies:

import matplotlib.pyplot as plt
fig, axes = plt.subplots(nrows=2)
axes[0].set_aspect(1.)
axes[0].plot(np.arange(10),np.arange(10))
axes[0].set_ylim([0,24])
axes[0].set_xlim([0,12])
axes[1].plot(np.arange(10),np.arange(10)*2.)
plt.show()

Works as I'd expect.  axes[0] gets shrunk in the x dimension to make the aspect ratio 1.

However:

import matplotlib.pyplot as plt
fig, axes = plt.subplots(nrows=2,sharex=True)
axes[0].set_aspect(1.)
axes[0].plot(np.arange(10),np.arange(10))
axes[0].set_ylim([0,24])
axes[0].set_xlim([0,12])
axes[1].plot(np.arange(10),np.arange(10)*2.)
plt.show()

does not work as I'd expect.  axes[0]'s ylim gets changed so that the line is no longer viewable (= 10-14).  In my opinion, the two calls should work the same, except in the second case, axes[1]'s xlim should be 0-12.  

Even worse, if I use the same aspect ratio in axes[1], they *both* do not show all the data: 

axes[0].set_aspect(1.)

It appears here that with sharex=True the shape of the axis no longer becomes mutable when set_aspect() is used, whereas if sharex=False set_aspect() can change the axis shape. I don't see any reason for sharex to foist this behaviour onto set_aspect().  

I guess the workaround is don't use sharex=True, but I actually think this is a bug.  

Thanks,   Jody

--
Jody Klymak    
http://web.uvic.ca/~jklymak/

On 2015/04/08 7:04 AM, Jody Klymak wrote:
> Maybe there is a reason for the default, but I really think the data
> view should be prioritized over the shape of the axis.

I forgot to include: I was trying to make everything sane (and 
reversible) under zoom and pan as well as reshaping and resizing.

On 2015/04/08 7:04 AM, Jody Klymak wrote:
> Following up on this, I’d like to complain about set_aspect()…
>
> If I do:
>
> import matplotlib.pyplot as plt
> fig, axes = plt.subplots(nrows=2,sharex=True)
> axes[0].set_ylim(0,1.)
> axes[0].set_aspect(1.)
> plt.show()
>
> the x-axis goes from 0. to 1., but axes[0]’s y-axis goes from 0.32 to
> 0.67.  Swapping the order of the y_lim call doesn’t help.  This is very
> un-intuitive, as I’d expect set_ylim() to set what data I see no matter
> what else is happening w/ the plot.
>
> I see that
> axes[0].set_aspect(1.,adjustable=‘box-forced’)
> will give the desired behaviour, but I really think it should be the
> default, not adjustable=‘datalim'.  I had no idea set_aspect() had this
> parameter until today, and have had several cursing matches with
> set_aspect as it kept changing my explicitly set data limits. set_ylim()
> should set the y limits.
>
> Just my opinion.  Maybe there is a reason for the default, but I really
> think the data view should be prioritized over the shape of the axis.

Jody,

I'm the guilty party for most of how set_aspect works.  I developed it a 
long time ago. Yes, there was a reason--still is, I'm 99% sure--but I 
don't remember everything, and don't want to take the time now to 
reconstruct the whole rationale.  When I was developing the behavior, I 
was paying a lot of attention to what happens under various scenarios of 
resizing and reshaping the window, and turning options on and off. 
There are some basic conflicts that arise when shared axes are combined 
with fixed aspect ratios, autoscaling, and gui-driven reshaping. 
Sometimes 'box-forced' does what people want, maybe more often than not; 
but I'm pretty sure it can lead to trouble, which is the reason it is 
not the default.

Eric

>
> Thanks,   Jody
>