matplotlib-users Mailing List for matplotlib

On 07/02/2012 12:36 PM, Pavol Juhas wrote:
> Done - https://github.com/matplotlib/matplotlib/issues/984
Thank you.
> I guess this means there is currently no way around it, right?
None that I know of offhand.
Eric
>
> Pavol
>
> ------------------------------------------------------------------------------
> Live Security Virtual Conference
> Exclusive live event will cover all the ways today's security and
> threat landscape has changed and how IT managers can respond. Discussions
> will include endpoint security, mobile security and the latest in malware
> threats. http://www.accelacomm.com/jaw/sfrnl04242012/114/50122263/
> _______________________________________________
> Matplotlib-users mailing list
> Mat...@li...
> https://lists.sourceforge.net/lists/listinfo/matplotlib-users
>

Done - https://github.com/matplotlib/matplotlib/issues/984
I guess this means there is currently no way around it, right?
Pavol

On Sun, Jul 1, 2012 at 12:50 PM, surfcast23 <sur...@gm...> wrote:
>
> Hi,
> I am translating a Matlab code to python and get the following error when
> the codes reaches the plotting section
>
> Warning (from warnings module):
> File "C:\Documents and Settings\My Documents\PHYSICS\Wave-eqn.py", line
> 40
> w = (D*v)
> RuntimeWarning: overflow encountered in multiply
> Traceback (most recent call last):
> File "C:\Documents and Settings\My Documents\PHYSICS\Wave-eqn\.py", line
> 50, in <module>
> ax.plot_wireframe(x,tdata,data, rstride=10, cstride=10)
> File "C:\Python32\lib\site-packages\mpl_toolkits\mplot3d\axes3d.py", line
> 906, in plot_wireframe
> tylines = [tY[i] for i in cii]
> File "C:\Python32\lib\site-packages\mpl_toolkits\mplot3d\axes3d.py", line
> 906, in <listcomp>
> tylines = [tY[i] for i in cii]
> IndexError: index out of bounds
>
> My code
>
> import numpy as np
> from numpy import *
> from math import pi
> from scipy.linalg import toeplitz
> from scipy.special import cotdg
> from mpl_toolkits.mplot3d import axes3d
> import matplotlib.pyplot as plt
>
>
>
> N = 512
> h = 2*np.pi/N
> x = h*(np.arange(N) + 1)
> t = 0
> dt = h / 4
> a = .1
> tmax = 15;
> tplot = .15;
> nplots = int(round((tmax/tplot)));
> plotgap = int(around(tplot/dt));
> c = a + np.sin(x - 1)**2
> v = np.exp(-100 * (x - 1)**2)
> vold = np.exp(-100 * (x - a*dt - 1)**2)
>
> #i = np.arange(1, N)
> #column = np.hstack([0, .5 * (-1**i) * cotdg(i * h/2)])
> #D = toeplitz(column, -column)
>
> column = ((0.5*(-1)**arange(1,N+1))*cotdg(arange(1,N+1))*(h/2));
> D = toeplitz(column,-column);print(D.shape);
>
> k = np.zeros(((nplots,N))); print(v.shape);print(k.shape);
> data = np.concatenate((v.reshape((512,1)).transpose(), k))#data =
> np.concatenate((v, k),axis = 1);
> #data = np.vstack([v,k]);
> tdata = t;
>
> for i in range(1,nplots+1):
> for n in range(1,plotgap+1):
> t = t+dt
> w = (D*v)
> vnew = vold-2*dt*c*w
> vold = v
> v = vnew
> data[i,:] = v[0,:]
> tdata = vstack([tdata, t])
>
> fig = plt.figure()
> ax = fig.add_subplot(111, projection='3d')
> #X, Y, Z = axes3d.get_test_data(0.05)
> ax.plot_wireframe(x,tdata,data, rstride=10, cstride=10)
>
> plt.show()
>
> I looked at the error line and it seems as if the y axes is where the
> problem is, but I am not seeing why and would appreciate any help. Thank
> you!
>
numpy arrays are indexed starting at 0, not 1. So when you populate your
"data" array with "data[i,:] = v[0,:]", and "i" only goes from 1 to nplots,
data[0,:] is left completely uninitialized (unless it is being done by some
of your pre-for-loop code, which is confusing to understand.)
What I can tell you is that the error isn't in plot_wireframe() as much as
the error exist with the inputs to plot_wireframe(). Perhaps the shapes
aren't right or something. I will try and look at your code closer
tomorrow and see if I can figure it out, but I suggest double-checking
those arrays.
Cheers!
Ben Root

On 06/27/2012 09:12 PM, Mads Ipsen wrote:
> Hi,
>
> Suppose you do this:
>
> axes = self.figure().get_axes()
> contour = axes.contourf(x,y,z)
> colorbar = self.figure().colorbar(contour)
>
> Suppose that the contour data changes, can you update the colorbar with
> the new data?
>
> Currently I remove the colorbar and insert a new one - but I have a
> feeling that something smarter could be done.
When you say "the contour data changes", I assume you mean you are 
contouring a new set of data, with new contour levels, and not just 
changing the cmap. In this case there is no point in trying to save 
something from the old colorbar, so remaking it is the right thing to 
do. Simple, effective, foolproof. (If only the cmap is changed, then 
the existing colorbar should be getting updated automatically.)
Eric
>
> Best regards,
>
> Mads
>

Mads,
On Thu, Jun 28, 2012 at 3:12 AM, Mads Ipsen <mad...@gm...> wrote:
> Hi,
>
> Suppose you do this:
>
> axes = self.figure().get_axes()
> contour = axes.contourf(x,y,z)
> colorbar = self.figure().colorbar(contour)
>
> Suppose that the contour data changes, can you update the colorbar with
> the new data?
>
> Currently I remove the colorbar and insert a new one - but I have a
> feeling that something smarter could be done.
>
> Best regards,
>
> Mads
>
The only way I could think of to do what you want a bit more intelligently
is to reuse the existing QuadContourMesh that comes from the initial call
to contourf. Updates to that "ScalarMappable" object should then trigger
updates to the existing colorbar. Not entirely sure which code would be
cleaner, though, because reusing a QuadContourMesh isn't easy.
Cheers!
Ben Root

Mads,
On Thu, Jun 28, 2012 at 7:47 AM, Mads Ipsen <mad...@gm...> wrote:
> I want to make some mods to a few selected tick labels.
>
> For example, if I do
>
> label = axes.yaxis.get_major_ticks()[2].label
> label.set_fontsize(size)
> label.set_rotation('vertical')
>
> the font size and the orientation of the tick label is changed. However,
> if try
>
> label.set_text('Foo')
>
> the tick label is *not* modified.
>
> Any clues?
>
> Best regards,
>
> Mads
>
>
I can't seem to reproduce your issue. If I call "label.get_text()" after
calling "label.set_text('Foo')", I get 'Foo' back. Did you mean that the
text on the plot was not updated? If so, I wouldn't be totally surprised.
Which version of matplotlib (and backend) are you using?
Cheers!
Ben Root

On 07/02/2012 06:34 AM, Pavol Juhas wrote:
> Hi all,
>
> I have been using backend-neutral matplotlib (0.99) widgets to implement
> a simple GUI control, which can be opened from IPython pylab session.
> The widgets are all contained in a dedicated figure, however if this
> figure is the current one, pyplot commands draw through the widget
> elements.
>
> Is there any option to lock a matplotlib figure and thus protect it
> from pyplot commands? If not, are there any plans to add such
> function? I think this would surely make the matplotlib widgets more
> useful.
>
> Thank you,
>
> Pavol
>
> PS: Actually, the built-in subplot_tool function seems to suffer from
> the same issue - here is an example which draws a line through the
> Reset button:
>
> $ ipython -pylab
> In [1]: subplot_tool()
> # click any slider
> In [2]: plot(range(2))
This looks like a bug. Would you report it on Github as a New Issue, 
please, noting your mpl version and backend?
https://github.com/matplotlib/matplotlib/issues.
I can replicate it with 1.2.x on Linux.
Thank you.
Eric
>
> ------------------------------------------------------------------------------
> Live Security Virtual Conference
> Exclusive live event will cover all the ways today's security and
> threat landscape has changed and how IT managers can respond. Discussions
> will include endpoint security, mobile security and the latest in malware
> threats. http://www.accelacomm.com/jaw/sfrnl04242012/114/50122263/
> _______________________________________________
> Matplotlib-users mailing list
> Mat...@li...
> https://lists.sourceforge.net/lists/listinfo/matplotlib-users
>

Hi all,
I have been using backend-neutral matplotlib (0.99) widgets to implement
a simple GUI control, which can be opened from IPython pylab session.
The widgets are all contained in a dedicated figure, however if this
figure is the current one, pyplot commands draw through the widget
elements.
Is there any option to lock a matplotlib figure and thus protect it
from pyplot commands? If not, are there any plans to add such
function? I think this would surely make the matplotlib widgets more
useful.
Thank you,
Pavol
PS: Actually, the built-in subplot_tool function seems to suffer from
 the same issue - here is an example which draws a line through the
 Reset button:
$ ipython -pylab
In [1]: subplot_tool()
# click any slider
In [2]: plot(range(2))

Hi all!
How can I use the keyboard events up/down/right/left without starting an interaction with the navigation toolbar?
I connected to the keypress events and everything looks ok until I press the down key: my connected method gets called once and the navigation toolbar then gets the focus. From that point on it catches all of my arrow presses.
I noticed that using shifted keys could be a workaround but I really do not like it...
Thanks in advance for help!
cheers JF
-
Helmholtz Zentrum Geesthacht
Institut für Werkstoffforschung
Abteilung WPN, Instrument REFSANS
Lichtenbergstr. 1
85747 Garching FRM II
Tel.: +49 (0)89 289 10762
Internet: http://www.frm2.tum.deHelmholtz-Zentrum Geesthacht 
Zentrum für Material- und Küstenforschung GmbH 
Max-Planck-Straße 1 I 21502 Geesthacht I Deutschland/Germany 
Geschäftsführer/Board of Management: Prof. Dr. Wolfgang Kaysser, Dipl.-Ing. Michael Ganß 
Vorsitzender des Aufsichtsrates/Chairman of the Supervisory Board: MinDirig Wilfried Kraus 
Amtsgericht Lübeck HRB 285 GE (Register Court) 
Internet: http://www.hzg.de