matplotlib-users Mailing List for matplotlib

Hi:

I am getting incorrect renderings when using \hat{x} or \vec{x}. The
following code

#####################################################
import matplotlib.pylab as plt

plt.axes([0.1, 0.15, 0.8, 0.75])
plt.plot(range(10))

plt.xlabel(r'$\hat{y}$  $\vec{x}$ $x^2 + y^2$', fontsize=20)
plt.show()
#####################################################

produce the attached plot. Note that the "hat" and the "arrow" are in
the wrong place. The other Latex part looks OK.

I am running version 1.2.x (built from commit
11e528425e230a3e23d04202aea23d88d40d9c4c) and Ubuntu 11.10.

Any ideas about how to solve the problem?

Alejandro.

thanks for the responses.
Sebastians suggestion to use tissot function is exactly what I needed.
map.tissot(lon, lat, r, 96)
thanks again

On Wed, Dec 7, 2011 at 6:36 PM, Thomas Lecocq <thl...@ms...> wrote:

>  Dear,
>
> you can try my tutorial to achieve this properly :
> http://www.geophysique.be/2011/02/20/matplotlib-basemap-tutorial-09-drawing-circles/
>
> Cheers,
>
> Thom
>
> ps : on the "things to do when I have some time" list : commit a method to
> the default basemap package to do this...
>
>
> ------------------------------
> Date: Tue, 6 Dec 2011 21:23:18 -0600
> From: ben...@ou...
> To: que...@gm...
> CC: Mat...@li...
> Subject: Re: [Matplotlib-users] add cirlce around lat lon
>
>
>
>
> On Tuesday, December 6, 2011, questions anon <que...@gm...>
> wrote:
> > I would like to draw a simple circle around a specified latitude and
> longitude but I cannot find an appropriate command.
> > I have tried using
> > map.drawgreatcircle(myLON, myLAT,myLON, myLAT, linewidth=20,color='k')
> > but this doesn't do anything
> > or even
> > map.drawgreatcircle(myLON+1, myLAT+1,myLON-1, myLAT-1,
> linewidth=2,color='k')
> > and this appears to draw a line.
> > Any other commands I could try for this?
> > thanks in advance
> >
>
> drawgreatcircle() doesn't actually draw a circle, but rather an arc that
> represents the shortest distance between two points on the globe.
>
> Maybe you would rather use a Circle object?
>
> Ben Root
> ------------------------------------------------------------------------------
> Cloud Services Checklist: Pricing and Packaging Optimization This white
> paper is intended to serve as a reference, checklist and point of
> discussion for anyone considering optimizing the pricing and packaging
> model of a cloud services business. Read Now!
> http://www.accelacomm.com/jaw/sfnl/114/51491232/
> _______________________________________________ Matplotlib-users mailing
> list Mat...@li...
> https://lists.sourceforge.net/lists/listinfo/matplotlib-users
>
>
> ------------------------------------------------------------------------------
> Cloud Services Checklist: Pricing and Packaging Optimization
> This white paper is intended to serve as a reference, checklist and point
> of
> discussion for anyone considering optimizing the pricing and packaging
> model
> of a cloud services business. Read Now!
> http://www.accelacomm.com/jaw/sfnl/114/51491232/
> _______________________________________________
> Matplotlib-users mailing list
> Mat...@li...
> https://lists.sourceforge.net/lists/listinfo/matplotlib-users
>
>

Wed, 7 Dec 2011 20:29:22 +0100
Jérôme a écrit:

> Is there a way to automatically resize the axis and nicely center the whole
> set {axes + ticklabels + labels} in the figure ?
> 
> One could use add_axes and play with the coordinates until he gets something
> nice, but it gets complicated to have it automatic as things depends on
> - the number of digits of y-axis ticklabels
> - whether or not a secundary y-axis is added on the right (using twinx)

Hi again, sorry for multi-posting.

Apparently, figure.tight_layout() does not take into account the secondary
y-axis on the right.

Is this a known limitation ? (I don't see it on the caveats paragraph [1].)

Or is this the use I make of it that is incorrect ?

Example :

--------------------------------------------------------------------------

import pylab

fig = pylab.figure()

data_1 = [0,1,2,3]
data_2 = [0,5,250,30000]

lines = []

# Primary axis
ax1 = fig.add_subplot(1,1,1)
lines.extend (ax1.plot(data_1, 'b'))

# Secondary axis
ax2 = pylab.twinx(ax1)
lines.extend (ax2.plot(data_2, 'g'))

labels = ['Data 1', 'Data 2']

fig.tight_layout()

pylab.show()


--------------------------------------------------------------------------

Thanks.

[1] http://matplotlib.sourceforge.net/users/tight_layout_guide.html

-- 
Jérôme

On Wed, Dec 7, 2011 at 11:08 AM, claudius <cla...@ya...> wrote:

>
> I would like to draw a round pie in a rectangle figure. At the moment I'm
> using something like:
>
>        fig = plt.figure( figsize = figsize, dpi=inch)
>
>        # plot actually
>        ax = fig.add_subplot( 1, 1, 1 )
>        ax.pie( value_list, labels = labels_list, **kwargs )
>
>        plt.savefig( plt_pathname )
>        plt.close()
>
> If the figsize is not square ( eg. [4, 4]) then the resulting figure will
> be
> stretched, elipsoid.
> Can I overcome this issue.
> Thanks in advance
>
> You can set the aspect of the axes object:

ax.set_aspect('equal')

Best,
-Tony

Hi again.

Wed, 7 Dec 2011 20:29:22 +0100
Jérôme a écrit:

> Is there a way to automatically resize the axis and nicely center the whole
> set {axes + ticklabels + labels} in the figure ?

[...]
 
> Or did I miss something ?

It seems I missed figure.tight_layout().

Sorry about that...

-- 
Jérôme

Hi all.

The position of an axes is fixed at creation, regardless of the what goes
outside the plot area. If the numbers on the y-axis are big enough (say, 7
digits) and a label is added, the label gets out of the figure.

Example :

--------------------------------------------------------------------------

import pylab

data = [0,1,2,3000000]

fig = pylab.figure()

ax1 = fig.add_subplot(1,1,1)
ax1.plot(data)
ax1.set_ylabel('label_axis_y1')

pylab.show()

--------------------------------------------------------------------------

Is there a way to automatically resize the axis and nicely center the whole
set {axes + ticklabels + labels} in the figure ?

One could use add_axes and play with the coordinates until he gets something
nice, but it gets complicated to have it automatic as things depends on
- the number of digits of y-axis ticklabels
- whether or not a secundary y-axis is added on the right (using twinx)

Or did I miss something ?

Thanks.

-- 
Jérôme

Thank you very much!  I was trying to do something like this in
legend_handler.py but this is such a simple fix!

Best,
  Adrian.

On Wed, Dec 7, 2011 at 9:11 AM, Jae-Joon Lee <lee...@gm...> wrote:
> I just pushed a change that I believe fixes this problem
>
> https://github.com/matplotlib/matplotlib/commit/96caca8dd48d08e3106337ecdeae82fa0236b86b
>
> Required change is very minor, so you may apply the patch by yourself.
> If you need a workaround, let me know.
> Regards,
>
> -JJ
>
>
> On Tue, Dec 6, 2011 at 11:46 AM, Adrian Del Maestro <ag...@gm...> wrote:
>> Hello,
>>
>> I have run across an inconsistency in the zorder of markers and lines
>> in the legend for an errorbar as opposed to a plot in matplotlib
>> v1.1.0.  After some considerable amount of time reading
>> legend_handler.py and the information at
>> http://matplotlib.sourceforge.net/users/legend_guide.html I was unable
>> to figure out how to make the errorbar markers sit 'on top' of the
>> lines in a legend.
>>
>> For example the following code:
>>
>> import pylab as pl
>> import numpy as np
>>
>> x = pl.arange(-2,2,0.1)
>> y = x**2
>> dy = np.random.random(len(x))
>>
>> pl.figure(1)
>> pl.plot(x,y, color='k', linewidth=0.5, linestyle='-', marker='o',
>>        markerfacecolor='lime', markeredgecolor='k', markeredgewidth=0.5,
>> label='line1', markersize=10)
>> pl.legend(loc='best', frameon=False)
>>
>> pl.figure(2)
>> pl.errorbar(x,y,yerr=dy, color='k', linewidth=0.5, linestyle='-', marker='o',
>>        markerfacecolor='lime', markeredgecolor='k', markeredgewidth=0.5,
>>        label='line1', markersize=10, ecolor='lime', capsize=10)
>> pl.legend(loc='best', frameon=False)
>>
>> pl.show()
>>
>> produces a legend for the plot (figure(1)) with the markers on top of
>> the lines, but the legend for the errorbar (figure(2)) has this
>> reversed.
>>
>> Any ideas/suggestions would be greatly appreciated.
>>
>> Best,
>>  Adrian.
>>
>> ------------------------------------------------------------------------------
>> Cloud Services Checklist: Pricing and Packaging Optimization
>> This white paper is intended to serve as a reference, checklist and point of
>> discussion for anyone considering optimizing the pricing and packaging model
>> of a cloud services business. Read Now!
>> http://www.accelacomm.com/jaw/sfnl/114/51491232/
>> _______________________________________________
>> Matplotlib-users mailing list
>> Mat...@li...
>> https://lists.sourceforge.net/lists/listinfo/matplotlib-users

I would like to draw a round pie in a rectangle figure. At the moment I'm
using something like:

        fig = plt.figure( figsize = figsize, dpi=inch)

        # plot actually
        ax = fig.add_subplot( 1, 1, 1 )
        ax.pie( value_list, labels = labels_list, **kwargs )

        plt.savefig( plt_pathname )
        plt.close()

If the figsize is not square ( eg. [4, 4]) then the resulting figure will be
stretched, elipsoid.
Can I overcome this issue.
Thanks in advance
-- 
View this message in context: http://old.nabble.com/Round-pie-in-non-square-figure-size-tp32929787p32929787.html
Sent from the matplotlib - users mailing list archive at Nabble.com.

Can you post an standalone example?
Maybe you want to set the *annotation_clip* parameter to False?

http://matplotlib.sourceforge.net/api/pyplot_api.html#matplotlib.pyplot.annotate

Regards,

-JJ


On Fri, Dec 2, 2011 at 10:19 PM, Neal Becker <ndb...@gm...> wrote:
> Using horizontalalignment='right', it seems that if a point lies on the right
> edge of the plot, the annotation does not appear, even though (since the text
> should be right aligned), the text would have been on the plot and be visible.
>
> Any workaround?
>
>
> ------------------------------------------------------------------------------
> All the data continuously generated in your IT infrastructure
> contains a definitive record of customers, application performance,
> security threats, fraudulent activity, and more. Splunk takes this
> data and makes sense of it. IT sense. And common sense.
> http://p.sf.net/sfu/splunk-novd2d
> _______________________________________________
> Matplotlib-users mailing list
> Mat...@li...
> https://lists.sourceforge.net/lists/listinfo/matplotlib-users

I just pushed a change that I believe fixes this problem

https://github.com/matplotlib/matplotlib/commit/96caca8dd48d08e3106337ecdeae82fa0236b86b

Required change is very minor, so you may apply the patch by yourself.
If you need a workaround, let me know.
Regards,

-JJ


On Tue, Dec 6, 2011 at 11:46 AM, Adrian Del Maestro <ag...@gm...> wrote:
> Hello,
>
> I have run across an inconsistency in the zorder of markers and lines
> in the legend for an errorbar as opposed to a plot in matplotlib
> v1.1.0.  After some considerable amount of time reading
> legend_handler.py and the information at
> http://matplotlib.sourceforge.net/users/legend_guide.html I was unable
> to figure out how to make the errorbar markers sit 'on top' of the
> lines in a legend.
>
> For example the following code:
>
> import pylab as pl
> import numpy as np
>
> x = pl.arange(-2,2,0.1)
> y = x**2
> dy = np.random.random(len(x))
>
> pl.figure(1)
> pl.plot(x,y, color='k', linewidth=0.5, linestyle='-', marker='o',
>        markerfacecolor='lime', markeredgecolor='k', markeredgewidth=0.5,
> label='line1', markersize=10)
> pl.legend(loc='best', frameon=False)
>
> pl.figure(2)
> pl.errorbar(x,y,yerr=dy, color='k', linewidth=0.5, linestyle='-', marker='o',
>        markerfacecolor='lime', markeredgecolor='k', markeredgewidth=0.5,
>        label='line1', markersize=10, ecolor='lime', capsize=10)
> pl.legend(loc='best', frameon=False)
>
> pl.show()
>
> produces a legend for the plot (figure(1)) with the markers on top of
> the lines, but the legend for the errorbar (figure(2)) has this
> reversed.
>
> Any ideas/suggestions would be greatly appreciated.
>
> Best,
>  Adrian.
>
> ------------------------------------------------------------------------------
> Cloud Services Checklist: Pricing and Packaging Optimization
> This white paper is intended to serve as a reference, checklist and point of
> discussion for anyone considering optimizing the pricing and packaging model
> of a cloud services business. Read Now!
> http://www.accelacomm.com/jaw/sfnl/114/51491232/
> _______________________________________________
> Matplotlib-users mailing list
> Mat...@li...
> https://lists.sourceforge.net/lists/listinfo/matplotlib-users

Dear,





you can try my tutorial to achieve this properly : 
http://www.geophysique.be/2011/02/20/matplotlib-basemap-tutorial-09-drawing-circles/





Cheers,





Thom





ps : on the "things to do when I have some time" list : commit a method to the default basemap package to do this...


Date: Tue, 6 Dec 2011 21:23:18 -0600
From: ben...@ou...
To: que...@gm...
CC: Mat...@li...
Subject: Re: [Matplotlib-users] add cirlce around lat lon



On Tuesday, December 6, 2011, questions anon <que...@gm...> wrote:
> I would like to draw a simple circle around a specified latitude and longitude but I cannot find an appropriate command.

> I have tried using
> map.drawgreatcircle(myLON, myLAT,myLON, myLAT, linewidth=20,color='k')
> but this doesn't do anything
> or even
> map.drawgreatcircle(myLON+1, myLAT+1,myLON-1, myLAT-1, linewidth=2,color='k')

> and this appears to draw a line.
> Any other commands I could try for this?
> thanks in advance
>

drawgreatcircle() doesn't actually draw a circle, but rather an arc that represents the shortest distance between two points on the globe.


Maybe you would rather use a Circle object?

Ben Root

------------------------------------------------------------------------------
Cloud Services Checklist: Pricing and Packaging Optimization
This white paper is intended to serve as a reference, checklist and point of 
discussion for anyone considering optimizing the pricing and packaging model 
of a cloud services business. Read Now!
http://www.accelacomm.com/jaw/sfnl/114/51491232/
_______________________________________________
Matplotlib-users mailing list
Mat...@li...
https://lists.sourceforge.net/lists/listinfo/matplotlib-users 		 	   		  

>>> Benjamin Root  12/07/11 4:16 AM >>>
>
> Use draw_idle() if performance is an issue.  Also, you don't have to
redraw
> everything. You can save the object returned by avline() 


> and in subsequent draws, just modify the data. Usually, there is a
set_data()
>  or a set_xy() method you can use for this.




I didn't find those methods for the axvline. I thought it would have
them like 
other Line2D objects.


This now works, though it sometimes creates ghost double lines, and
isn't the 
most performant solution:


   def update_marker(self, event):
      print "plotMarker() pos = ", event.xdata       # DEBUG
      self.marker=self.ax1
      
      #print 'button=%d, x=%d, y=%d, xdata=%f, ydata=%f'%(
      #   event.button, event.x, event.y, event.xdata, event.ydata)    
# DEBUG      
      
      # Create markers (vertical lines) in both plots
      self.marker=self.ax1.axvline(x=event.xdata, linewidth=1.5,
color='r')
      self.marker=self.ax2.axvline(x=event.xdata, linewidth=1.5,
color='r')
 
      # We need to remove all unnecessary marker in plot 1 and plot 2
      # TODO: find better method, keeps double marker sometimes
      for i in range(1, len(self.ax1.lines)-1):
        self.ax1.lines[i].remove()
      for i in range(1, len(self.ax2.lines)-1):
        self.ax2.lines[i].remove()  
        
      self.marker=self.ax1.axvline(x=event.xdata, linewidth=1.5,
color='r')
      self.marker=self.ax1.axvline(x=event.xdata, linewidth=1.5,
color='r')
      self.canvas.draw_idle()





Thanks for the help.




Cheers,




Sven

On Tuesday, December 6, 2011, questions anon <que...@gm...>
wrote:
> I would like to draw a simple circle around a specified latitude and
longitude but I cannot find an appropriate command.
> I have tried using
> map.drawgreatcircle(myLON, myLAT,myLON, myLAT, linewidth=20,color='k')
> but this doesn't do anything
> or even
> map.drawgreatcircle(myLON+1, myLAT+1,myLON-1, myLAT-1,
linewidth=2,color='k')
> and this appears to draw a line.
> Any other commands I could try for this?
> thanks in advance
>

drawgreatcircle() doesn't actually draw a circle, but rather an arc that
represents the shortest distance between two points on the globe.

Maybe you would rather use a Circle object?

Ben Root

On Tuesday, December 6, 2011, Sven Duscha <du...@as...> wrote:
> Hi,
>
>
> I kind of got the basic functionality working using
>
>
>      cid = self.fig.canvas.mpl_connect('motion_notify_event',
> self.update_marker)
>
>
>   def update_marker(self, event):
>      print "plotMarker()"       # DEBUG
>      self.marker=self.ax1
>
>      #print 'button=%d, x=%d, y=%d, xdata=%f, ydata=%f'%(
>      #   event.button, event.x, event.y, event.xdata, event.ydata)
> # DEBUG
>
>      self.ax1.axvline(x=event.xdata, linewidth=1, color='r')
>      self.canvas.draw()  # redraw
>
>
>
>
>
> but I still have the problem that previous lines still appear on the
> plot.
>
>
> Also, canvas.draw() seems to be a bit of a performance hog. Does anyone
> have
> any more sophisticated solution to this?
>
>
> Cheers,
>
>
> Sven
>

Use draw_idle() if performance is an issue.  Also, you don't have to redraw
everything. You can save the object returned by avline() and in subsequent
draws, just modify the data. Usually, there is a set_data() or a set_xy()
method you can use for this.

Ben Root

Hi,


I kind of got the basic functionality working using


      cid = self.fig.canvas.mpl_connect('motion_notify_event',
self.update_marker)


   def update_marker(self, event):
      print "plotMarker()"       # DEBUG
      self.marker=self.ax1
      
      #print 'button=%d, x=%d, y=%d, xdata=%f, ydata=%f'%(
      #   event.button, event.x, event.y, event.xdata, event.ydata)    
# DEBUG      
      
      self.ax1.axvline(x=event.xdata, linewidth=1, color='r')
      self.canvas.draw()  # redraw





but I still have the problem that previous lines still appear on the
plot.


Also, canvas.draw() seems to be a bit of a performance hog. Does anyone
have 
any more sophisticated solution to this?


Cheers,


Sven

I would like to draw a simple circle around a specified latitude and
longitude but I cannot find an appropriate command.
I have tried using
map.drawgreatcircle(myLON, myLAT,myLON, myLAT, linewidth=20,color='k')
but this doesn't do anything
or even
map.drawgreatcircle(myLON+1, myLAT+1,myLON-1, myLAT-1,
linewidth=2,color='k')
and this appears to draw a line.
Any other commands I could try for this?
thanks in advance

Hi,




how could I realize a moving marker line under the mouse pointer?


I know about event handling
http://matplotlib.sourceforge.net/users/event_handling.html


and would use ‘ axes_enter_event’ to react on, then draw a line 
with


axvline(x=event.xdata, linewidth=1, color='r')


but how can I delete the line again without deleting my whole plot 
I have on that axes/subplot?


Also I didn't see any "mouse move" event that would recognize a 
change in cursor position? Or is that what is meant with 
'motion_notify_event'?


If that is so, it comes down to know how to delete a vertical line on 
a mouse_notify_event. Is there a way to get a (named) list of plots 
on a particular subplot? 


In a different context I tried to keep my on list  in a python 
self.axes list, but that was very tedious and I didn't finish 
it to work properly.




Cheers,




Sven

>>> Sven Duscha 12/06/11 4:50 PM >>>

On Dec 6, 2011, at 4:34 PM, Till Stensitzki wrote:
> I think you have to plot something, else matplotlib don't know where
to draw the 
> ticks.



Thankfully that got solved; I had an additional


self.ax1.get_xaxis().set_visible(False)


somewhere in my code. Got a bit too convoluted.



Cheers,


Sven

-- 
Sven Duscha
ASTRON
P.O. Box 2
7990 AA,  Dwingeloo, The Netherlands
Phone: +31 521 595 241
Email: du...@as...