matplotlib-users Mailing List for matplotlib

On 07/26/2010 12:17 PM, David Mashburn wrote:
> Ben and Eric,
>
> Thanks so much for your help!
>
> I'm trying to turn change some of the rcParams in my script... Here is a
> test of what happens:
>
> import matplotlib
> matplotlib.rcParams['keymap.fullscreen']=''
>
> Traceback (most recent call last):
>    File "<input>", line 1, in<module>
>    File "/usr/lib/pymodules/python2.6/matplotlib/__init__.py", line 623,
> in __setitem__
>      See rcParams.keys() for a list of valid parameters.'%key)
> KeyError: 'keymap.fullscreen is not a valid rc parameter.See
> rcParams.keys() for a list of valid parameters.'
>
> It seems that none of the keymap parameters are defined in my rcParams
> dictionary...
>
> I also tried adding some of the "keymap" lines out of the online rc file:
> http://matplotlib.sf.net/_static/matplotlibrc
> to my rc file and it told me:   Bad key "keymap.pan"
>
> I've got version 0.99.0 on ubuntu.

Aha!  I thought you were using 1.0.  For 1.0, these things are rc 
settings; I had no idea they even existed back in 0.99.0.  I only 
tripped over "f" very recently.

The key bindings are coded in the key_press() method of 
FigureManagerBase.  The callback is connected to the canvas in the 
__init__ method.  The trick is to disconnect the callback:

fig = figure()
fig.canvas.mpl_disconnect(3)
plot([1,2,3])

Now key presses have no effect in that figure.  The bad thing here is 
that I used the cid 3, a seemingly random number.  I suspect, though, 
that once you find out what it is in your version of mpl (and it may 
still be 3), you will be able to rely on it for your purposes.  To find 
it, print out fig.canvas.callbacks.callbacks and look for 
key_press_event.  You could have your program use this dictionary to 
look it up.


Eric
>
> Is this a new feature?  Do you know what I can try instead??
>
> Thanks!
> -David
>
> On 07/26/2010 10:18 AM, Benjamin Root wrote:
>> On Mon, Jul 26, 2010 at 3:09 PM, David Mashburn
>> <dav...@gm...<mailto:dav...@gm...>  <mailto:dav...@gm...<mailto:dav...@gm...>>>  wrote:
>>
>>      Hello!
>>
>>      My name is David Mashburn, and I have been a very happy user of
>>      matplotlib for almost 5 years now, so let me start by saying thank you
>>      for this wonderful piece of software!  I has been a blessing to me!
>>
>>      I am creating a program that performs actions based on key commands in
>>      matplotlib, and I keep bumping into conflicts with the interactive
>>      keyboard navigation system.  Is there any way to bypass or disable some
>>      or all of the keybindings hooked to interactive navigation?
>>
>>      By the way, I'm using the WxAgg backend if that matters.
>>
>>      Thanks!
>>      -David
>>
>>
>> David,
>>
>> I have not tested this myself, but I believe that if you set
>> "navigate=False" as a keyword argument for making your axes, this will
>> stop the callbacks from being registered.
>
> That completely disables the toolbar, which is likely not what the OP wants.
>
> There are a bunch of rcParams, listed at the bottom of
> matplotlibrc_template (and of the default matplotlibrc file that is
> installed) that can be used to turn off the key bindings without
> disabling the toolbar.
>
> Personally, I think that putting in all those key bindings by default
> was a major mistake, one that should be reversed ASAP.  Having the plot
> jump to fullscreen when one bumps "f" is rude. The first time it
> happened I had no idea what was going on or how to recover from it--I
> thought maybe my X-server had gone bonkers. And does one actually save a
> displayed figure so often that it is useful to have the "s" shortcut to
> save one from having to click the toolbar button?  I could go on with
> this complaint...
>
> Eric
>
>>
>> Thanks for using matplotlib!
>>
>> Ben Root
>>
>
>
>
> ------------------------------------------------------------------------------
> The Palm PDK Hot Apps Program offers developers who use the
> Plug-In Development Kit to bring their C/C++ apps to Palm for a share
> of $1 Million in cash or HP Products. Visit us here for more details:
> http://ad.doubleclick.net/clk;226879339;13503038;l?
> http://clk.atdmt.com/CRS/go/247765532/direct/01/
> _______________________________________________
> Matplotlib-users mailing list
> Mat...@li...
> https://lists.sourceforge.net/lists/listinfo/matplotlib-users

Ben and Eric,

Thanks so much for your help!

I'm trying to turn change some of the rcParams in my script... Here is a 
test of what happens:

import matplotlib
matplotlib.rcParams['keymap.fullscreen']=''

Traceback (most recent call last):
  File "<input>", line 1, in <module>
  File "/usr/lib/pymodules/python2.6/matplotlib/__init__.py", line 623, 
in __setitem__
    See rcParams.keys() for a list of valid parameters.'%key)
KeyError: 'keymap.fullscreen is not a valid rc parameter.See 
rcParams.keys() for a list of valid parameters.'

It seems that none of the keymap parameters are defined in my rcParams 
dictionary...

I also tried adding some of the "keymap" lines out of the online rc file:
http://matplotlib.sf.net/_static/matplotlibrc
to my rc file and it told me:   Bad key "keymap.pan"

I've got version 0.99.0 on ubuntu.

Is this a new feature?  Do you know what I can try instead??

Thanks!
-David

On 07/26/2010 10:18 AM, Benjamin Root wrote:
> On Mon, Jul 26, 2010 at 3:09 PM, David Mashburn
> <dav...@gm... <mailto:dav...@gm...> <mailto:dav...@gm... <mailto:dav...@gm...>>> wrote:
>
>     Hello!
>
>     My name is David Mashburn, and I have been a very happy user of
>     matplotlib for almost 5 years now, so let me start by saying thank you
>     for this wonderful piece of software!  I has been a blessing to me!
>
>     I am creating a program that performs actions based on key commands in
>     matplotlib, and I keep bumping into conflicts with the interactive
>     keyboard navigation system.  Is there any way to bypass or disable some
>     or all of the keybindings hooked to interactive navigation?
>
>     By the way, I'm using the WxAgg backend if that matters.
>
>     Thanks!
>     -David
>
>
> David,
>
> I have not tested this myself, but I believe that if you set
> "navigate=False" as a keyword argument for making your axes, this will
> stop the callbacks from being registered.

That completely disables the toolbar, which is likely not what the OP wants.

There are a bunch of rcParams, listed at the bottom of 
matplotlibrc_template (and of the default matplotlibrc file that is 
installed) that can be used to turn off the key bindings without 
disabling the toolbar.

Personally, I think that putting in all those key bindings by default 
was a major mistake, one that should be reversed ASAP.  Having the plot 
jump to fullscreen when one bumps "f" is rude. The first time it 
happened I had no idea what was going on or how to recover from it--I 
thought maybe my X-server had gone bonkers. And does one actually save a 
displayed figure so often that it is useful to have the "s" shortcut to 
save one from having to click the toolbar button?  I could go on with 
this complaint...

Eric

>
> Thanks for using matplotlib!
>
> Ben Root
>

On 07/26/2010 10:13 AM, Nikolaus Rath wrote:
> Hello,
>
> I would like to draw a couple of contour plots. The plots are on
> separate figures, but they should all have exactly the same color
> mapping (i.e, the same Z value should correspond to the same color in
> all plots).
>
> What's the best way to achieve this?

I think that all you will need is to specify the same cmap (if you are 
not using the default) and contour levels for both plots.  You should 
not need to use the norm, unless you are doing something fancy; and in 
that case, all you need to do is specify the same norm for both plots. 
The real key, though, is explicitly specifying the set of contour levels 
that you want, and using the same one for both plots.

e.g.:

from pylab import * # yes, this is not recommended...
clevs = arange(0.3, 0.701, 0.1)
fig1 = figure()
contourf(rand(20,20), levels=clevs, extend='both')
colorbar()
fig2 = figure()
contourf(rand(30,40), levels=clevs, extend='both')
colorbar()
show()



You need to use the extend argument only if you want values beyond the 
levels you specify to constitute open-ended ranges.  For example, if you 
have data values from 0 to 10, but most are between 4 and 6, you could 
specify levels as numpy.arange(4, 6.01, 0.1) and use extend='both'. Then 
all values greater than 6 would get a color at one end, and all values 
less than 4 would get another color at the other end.  If you use a 
colorbar, it will see the extend argument that was used in contourf, and 
draw the colorbar with pointed ends to indicate "anything over" and 
"anything under".  As illustrated above.

Eric

>
>  From the documentation I gather that I should use the norm and extend
> keyword arguments, but I wasn't really able to figure out what values
> I have to pass and how the matplotlib.colors.Normalize instance works.
>
>
> Thanks,
>
>     -Nikolaus
>

On 07/26/2010 10:18 AM, Benjamin Root wrote:
> On Mon, Jul 26, 2010 at 3:09 PM, David Mashburn
> <dav...@gm... <mailto:dav...@gm...>> wrote:
>
>     Hello!
>
>     My name is David Mashburn, and I have been a very happy user of
>     matplotlib for almost 5 years now, so let me start by saying thank you
>     for this wonderful piece of software!  I has been a blessing to me!
>
>     I am creating a program that performs actions based on key commands in
>     matplotlib, and I keep bumping into conflicts with the interactive
>     keyboard navigation system.  Is there any way to bypass or disable some
>     or all of the keybindings hooked to interactive navigation?
>
>     By the way, I'm using the WxAgg backend if that matters.
>
>     Thanks!
>     -David
>
>
> David,
>
> I have not tested this myself, but I believe that if you set
> "navigate=False" as a keyword argument for making your axes, this will
> stop the callbacks from being registered.

That completely disables the toolbar, which is likely not what the OP wants.

There are a bunch of rcParams, listed at the bottom of 
matplotlibrc_template (and of the default matplotlibrc file that is 
installed) that can be used to turn off the key bindings without 
disabling the toolbar.

Personally, I think that putting in all those key bindings by default 
was a major mistake, one that should be reversed ASAP.  Having the plot 
jump to fullscreen when one bumps "f" is rude. The first time it 
happened I had no idea what was going on or how to recover from it--I 
thought maybe my X-server had gone bonkers. And does one actually save a 
displayed figure so often that it is useful to have the "s" shortcut to 
save one from having to click the toolbar button?  I could go on with 
this complaint...

Eric

>
> Thanks for using matplotlib!
>
> Ben Root
>

On Mon, Jul 26, 2010 at 3:09 PM, David Mashburn
<dav...@gm...>wrote:

> Hello!
>
> My name is David Mashburn, and I have been a very happy user of
> matplotlib for almost 5 years now, so let me start by saying thank you
> for this wonderful piece of software!  I has been a blessing to me!
>
> I am creating a program that performs actions based on key commands in
> matplotlib, and I keep bumping into conflicts with the interactive
> keyboard navigation system.  Is there any way to bypass or disable some
> or all of the keybindings hooked to interactive navigation?
>
> By the way, I'm using the WxAgg backend if that matters.
>
> Thanks!
> -David
>
>
David,

I have not tested this myself, but I believe that if you set
"navigate=False" as a keyword argument for making your axes, this will stop
the callbacks from being registered.

Thanks for using matplotlib!

Ben Root

Hello,

I would like to draw a couple of contour plots. The plots are on
separate figures, but they should all have exactly the same color
mapping (i.e, the same Z value should correspond to the same color in
all plots).

What's the best way to achieve this?

>From the documentation I gather that I should use the norm and extend
keyword arguments, but I wasn't really able to figure out what values
I have to pass and how the matplotlib.colors.Normalize instance works.


Thanks,

   -Nikolaus

-- 
 »Time flies like an arrow, fruit flies like a Banana.«

  PGP fingerprint: 5B93 61F8 4EA2 E279 ABF6  02CF A9AD B7F8 AE4E 425C

Problem solved.

When I set up python on the Windows machine, I simply copied over the directories and everything in them from my Linux machine, including all of the python bytecode *.pyc files. Trying to run a script with those *.pyc files present was causing the problem as they were compiled on my Linux machine originally. 

The matplotlib call works fine when I move the scripts to a clean folder.

Thanks

Alex 

From: ben...@ou...
Date: Mon, 26 Jul 2010 09:56:27 -0500
Subject: Re: [Matplotlib-users] Import matplotlib.pyplot fails on windows XP
To: eck...@ho...
CC: mat...@li...

On Sat, Jul 24, 2010 at 4:57 AM, eck naysmith <eck...@ho...> wrote:







Hello,

I've installed the following Python packages on a Windows XP machine:

Python 2.6.5

Python 2.6 numpy-1.4.1

Python 2.6 matplotlib-0.99.3 [installer - matplotlib-0.99.3.win32-py2.6]



Python and Numpy work correctly. Matplotlib also works and as a test I tried successfully the following on the python interpretor:

from matplotlib.pyplot import *
plot([1,2,3])
show()

A graph then appears and thus so far so good.



However, when I try to run a certain python script, a problem occurs when importing from matplotlib:




	
	
	
	

Traceback (most recent
call last):
  File
"C:\Pythoncode\Games\Bridge_war_2\unitload2.py", line 2, in
<module>
    from matplotlib.pyplot
import plot, ylabel,xlabel, show
  File
"C:\Python26\lib\site-packages\matplotlib\pyplot.py", line
78, in <module>
    new_figure_manager,
draw_if_interactive, show = pylab_setup()
  File
"C:\Python26\lib\site-packages\matplotlib\backends\__init__.py",
line 25, in pylab_setup
    globals(),locals(),[backend_name])
  File
"C:\Python26\lib\site-packages\matplotlib\backends\backend_tkagg.py",
line 7, in <module>
    import Tkinter as Tk,
FileDialog
  File
"C:\Python26\lib\lib-tk\FileDialog.py", line 12, in
<module>
    from Dialog import
Dialog
  File
"C:\Pythoncode\Games\Bridge_war_2\Dialog.py", line 2, in
<module>
    import  wx
ImportError: No module
named wx

The error is that your Windows XP box does not have the wxPython module installed for Windows.

http://www.wxpython.org/download.php



I hope that helps,

Ben Root
 		 	   		  
_________________________________________________________________
http://clk.atdmt.com/UKM/go/197222280/direct/01/
We want to hear all your funny, exciting and crazy Hotmail stories. Tell us now

Hello!

My name is David Mashburn, and I have been a very happy user of
matplotlib for almost 5 years now, so let me start by saying thank you
for this wonderful piece of software!  I has been a blessing to me!

I am creating a program that performs actions based on key commands in
matplotlib, and I keep bumping into conflicts with the interactive
keyboard navigation system.  Is there any way to bypass or disable some
or all of the keybindings hooked to interactive navigation?

By the way, I'm using the WxAgg backend if that matters.

Thanks!
-David

Is there a simple function call for this? And finding the distance of
a point to the plane?

-Mathew

On Mon, Jul 26, 2010 at 9:49 AM, Benjamin Root <ben...@ou...> wrote:

> On Sat, Jul 24, 2010 at 1:51 PM, Josh Lawrence <jos...@gm...>wrote:
>
>>  Hello,
>>
>> I looked on your website for the different line styles. In the
>> documentation for matplotlib.lines.line2D.set_linestyle, the dashed
>> linestyle is listed as '-' and not '--'. It it my understanding that
>> dashed should be '--'. If I'm incorrect, sorry for the noise.
>>
>> Cheers,
>>
>> --
>> Josh Lawrence
>> Ph.D. Student
>> Clemson University
>>
>>
>>
> Looks like the formatter for the online documentation took the double-dash
> to mean a "long line", much like how LaTeX does.  Anybody know how to make
> the double-dash appear as two distinct dashes?
>
>
> http://matplotlib.sourceforge.net/api/artist_api.html?highlight=set_linestyle#matplotlib.lines.Line2D.set_linestyle
>
> Ben Root
>
>
After some reading of sphinx documentation, it appears to be a bug with
sphinx (or actually, "smartypants") because it should not be doing this sort
of interpretation within a docstring.  Anyway, supposedly the workaround is
to put double backticks around the part that needs to be treated literally:
``'--'``.  I tried this out and built the docs locally and it works... sort
of.  The text that is surrounded by double backticks are getting a different
background color.  This doesn't look great to me.  Maybe someone else has a
thought?

Ben Root

On 07/26/2010 08:18 PM, Timothy Vickery wrote:
> Hi everyone,
>
> I'm new to matplotlib, and I would like to create a typical plot of a
> timecourse (markers connected by lines), with the markers varying in
> size and/or style according to a separate set of data that corresponds
> to each timepoint. However, my understanding is that plot() will not
> accept an array of marker sizes, only a single float value that is
> applied to all markers. On the other hand, scatter() does allow
> variably-sized markers (as in the scatter_demo2.py example). However,
> there does not seem to be a straightforward way to add lines to this
> plot connecting markers in sequence. Can anyone recommend a work-around
> to this issue, or point out if I've missed something in the documentation?
>
> Thanks,
> Tim
>

You could use plot and scatter on the same plot. See the attached example.

#-------------------------------------------

import numpy as np
import matplotlib.pyplot as plt

N = 100
x = np.linspace(-10.0,10.0,N)
y  = 50.0*np.exp(-x**2)
size = 100.0*np.abs(np.sin(x))
plt.plot(x,y)
plt.scatter(x,y,s=size)

plt.show()

#-------------------------------------------


Regards,
João Luís

On Mon, Jul 26, 2010 at 2:50 PM, Mathew Yeates <mat...@gm...> wrote:

> Hi
> Is there any way to let the user rotate a 3D plot? I don't see an
> example which does this.
>
> -Mathew
>
>
What do you mean?  By default, all 3D plots are rotatable by merely clicking
and draging the plot around.  Or are you referring to how to set camera view
points?

Ben Root

Hi
Is there any way to let the user rotate a 3D plot? I don't see an
example which does this.

-Mathew

From: ben...@ou...
Date: Mon, 26 Jul 2010 09:56:27 -0500
Subject: Re: [Matplotlib-users] Import matplotlib.pyplot fails on windows XP
To: eck...@ho...
CC: mat...@li...

On Sat, Jul 24, 2010 at 4:57 AM, eck naysmith <eck...@ho...> wrote:







Hello,

I've installed the following Python packages on a Windows XP machine:

Python 2.6.5

Python 2.6 numpy-1.4.1

Python 2.6 matplotlib-0.99.3 [installer - matplotlib-0.99.3.win32-py2.6]



Python and Numpy work correctly. Matplotlib also works and as a test I tried successfully the following on the python interpretor:

from matplotlib.pyplot import *
plot([1,2,3])
show()

A graph then appears and thus so far so good.



However, when I try to run a certain python script, a problem occurs when importing from matplotlib:




	
	
	
	

Traceback (most recent
call last):
  File
"C:\Pythoncode\Games\Bridge_war_2\unitload2.py", line 2, in
<module>
    from matplotlib.pyplot
import plot, ylabel,xlabel, show
  File
"C:\Python26\lib\site-packages\matplotlib\pyplot.py", line
78, in <module>
    new_figure_manager,
draw_if_interactive, show = pylab_setup()
  File
"C:\Python26\lib\site-packages\matplotlib\backends\__init__.py",
line 25, in pylab_setup
    globals(),locals(),[backend_name])
  File
"C:\Python26\lib\site-packages\matplotlib\backends\backend_tkagg.py",
line 7, in <module>
    import Tkinter as Tk,
FileDialog
  File
"C:\Python26\lib\lib-tk\FileDialog.py", line 12, in
<module>
    from Dialog import
Dialog
  File
"C:\Pythoncode\Games\Bridge_war_2\Dialog.py", line 2, in
<module>
    import  wx
ImportError: No module
named wx

The error is that your Windows XP box does not have the wxPython module installed for Windows.

http://www.wxpython.org/download.php



I hope that helps,

Ben Root


No calls are made anywhere for wxPython so I don't know why it is seeking it. But when I installed wxPython anyway just to see what would happen, the traceback then ends at the penultimate line shown above i.e 

"File
"C:\Python26\lib\lib-tk\FileDialog.py", line 12, in
<module>
    from Dialog import
Dialog"
On the Traceback, each line beggining with 'File "C:\Python26......." is a seperate error. The most significant error is the first one:
'Traceback (most recent
call last):
  File
"C:\Pythoncode\Games\Bridge_war_2\unitload2.py", line 2, in
<module>
    from matplotlib.pyplot
import plot, ylabel,xlabel, show'
I suspected the problems were arising from the use of the statement 'from numpy import foobar' overwriting some of the names used by matplotlib, so I changed the import statement to 'import numpy' to avoid any potential namespace clash and adjusted the code accordingly. This does nothing and the exact same Traceback error messages occur.
Regards

Alex



 		 	   		  
_________________________________________________________________
http://clk.atdmt.com/UKM/go/195013117/direct/01/
We want to hear all your funny, exciting and crazy Hotmail stories. Tell us now

Hi everyone,

I'm new to matplotlib, and I would like to create a typical plot of a
timecourse (markers connected by lines), with the markers varying in size
and/or style according to a separate set of data that corresponds to each
timepoint. However, my understanding is that plot() will not accept an array
of marker sizes, only a single float value that is applied to all markers.
On the other hand, scatter() does allow variably-sized markers (as in the
scatter_demo2.py example). However, there does not seem to be a
straightforward way to add lines to this plot connecting markers in
sequence. Can anyone recommend a work-around to this issue, or point out if
I've missed something in the documentation?

Thanks,
Tim

I did what you showed me but the image disappears

On Mon, Jul 26, 2010 at 2:06 PM, <PH...@ge...> wrote:

>  Replace this line:
>
>   ax.plot(f/1000, Sserie)
>
>
>
> With this line:
>
>   ax.plot(f/1000.0, Sserie)
>
>
>
> And tell us how things go. Python 2.6 distinguishes between integers and
> floats very strictly. Hence:
>
> In [1]: 20/1000
>
> Out[1]: 0
>
>
>
> In [2]: 20.0/1000.0
>
> Out[2]: 0.02
>
>
>
> See the difference?
>
> -paul
>
>
>
> *From:* Waléria Antunes David [mailto:wal...@gm...]
> *Sent:* Monday, July 26, 2010 10:01 AM
> *To:* Angus McMorland
>
> *Cc:* mat...@li...
> *Subject:* Re: [Matplotlib-users] Hz to KHz
>
>
>
> i don't understand..
>
> On Mon, Jul 26, 2010 at 1:57 PM, Angus McMorland <am...@gm...>
> wrote:
>
> On 26 July 2010 12:47, Waléria Antunes David <wal...@gm...>
> wrote:
> > My code like this in django: http://pastebin.com/nzM5jvuc
>
> Are you perhaps suffering from integer division? How about dividing by
> 1000.0, instead of 1000?
>
> Angus.
>
>
> > i'm lost
> >
> > On Mon, Jul 26, 2010 at 1:42 PM, <PH...@ge...> wrote:
> >>
> >> Make sure that you’re dividing by 1000.0 (as opposed to 1000) to avoid
> any
> >> integer nonsense and make sure to reset your axis limits.
> >>
> >>
> >>
> >> Posting a small code snippet would help us get you on the right track.
> >>
> >> -p
> >>
> >>
> >>
> >> From: Waléria Antunes David [mailto:wal...@gm...]
> >> Sent: Monday, July 26, 2010 9:32 AM
> >> To: Matthieu Brucher
> >> Cc: mat...@li...
> >> Subject: Re: [Matplotlib-users] Hz to KHz
> >>
> >>
> >>
> >> I know...I tried but I'm using django and also when divided by 1000 the
> >> image does not appear...
> >> I don't know what to do.....help me....
> >>
> >> On Mon, Jul 26, 2010 at 1:27 PM, Matthieu Brucher
> >> <mat...@gm...> wrote:
> >>
> >> Hi,
> >>
> >> You may just divide them by 1000?
> >>
> >> Matthieu
> >>
> >> 2010/7/26 Waléria Antunes David <wal...@gm...>:
> >>
> >> > Hello all,
> >> >
> >> > I need to format the values of graphic to KHz.....my values are in Hz
> >> > see at idle python it displays the values as: 3000 3050 3100 .... 3400
> ,
> >> > but
> >> > I need to go where it will be displayed KHz:  3.0 3.1
> >> >
> >> > can someone help me?
> >>
> >> >
> >> >
> ------------------------------------------------------------------------------
> >> > The Palm PDK Hot Apps Program offers developers who use the
> >> > Plug-In Development Kit to bring their C/C++ apps to Palm for a share
> >> > of $1 Million in cash or HP Products. Visit us here for more details:
> >> > http://ad.doubleclick.net/clk;226879339;13503038;l?
> >> > http://clk.atdmt.com/CRS/go/247765532/direct/01/
> >> > _______________________________________________
> >> > Matplotlib-users mailing list
> >> > Mat...@li...
> >> > https://lists.sourceforge.net/lists/listinfo/matplotlib-users
> >> >
> >> >
> >>
> >>
> >>
> >> --
> >> Information System Engineer, Ph.D.
> >> Blog: http://matt.eifelle.com
> >> LinkedIn: http://www.linkedin.com/in/matthieubrucher
> >>
> >>
> >
> >
> ------------------------------------------------------------------------------
> > The Palm PDK Hot Apps Program offers developers who use the
> > Plug-In Development Kit to bring their C/C++ apps to Palm for a share
> > of $1 Million in cash or HP Products. Visit us here for more details:
> > http://ad.doubleclick.net/clk;226879339;13503038;l?
> > http://clk.atdmt.com/CRS/go/247765532/direct/01/
> > _______________________________________________
> > Matplotlib-users mailing list
> > Mat...@li...
> > https://lists.sourceforge.net/lists/listinfo/matplotlib-users
> >
> >
>
>
>
> --
>
> AJC McMorland
> Post-doctoral research fellow
> Neurobiology, University of Pittsburgh
>
>
>

On Jul 26, 2010 at 02:01 PM -0300, Waléria Antunes David wrote:
> i don't understand..

Just divide through by '1000.' or '1000.0' (same thing).

If you write 3100/1000, you'll get '3' because you are doing integer math. 
If you write 3100/1000.0, you'll get '3.1' because you are doing float math.

On 26 July 2010 13:01, Waléria Antunes David <wal...@gm...> wrote:
> i don't understand..

python (version < 3, I think) uses integer division when given integer
values, i.e. 1/1000 = 0, at the expense of giving a less accurate
answer than we might expect. If you tried to plot a bunch of values
divided by 1000, they may well all be integer 0, and so you wouldn't
see any line (it would be at the x-axis). You would need to force
python to do floating point division: 1/1000.0 = 0.001; to do this you
need to make sure one of the numbers in the division is floating
point: the simplest way to do this is 1/1000.0 (or 1/1000. for short).

I hope that helps,

Angus.

> On Mon, Jul 26, 2010 at 1:57 PM, Angus McMorland <am...@gm...> wrote:
>>
>> On 26 July 2010 12:47, Waléria Antunes David <wal...@gm...>
>> wrote:
>> > My code like this in django: http://pastebin.com/nzM5jvuc
>>
>> Are you perhaps suffering from integer division? How about dividing by
>> 1000.0, instead of 1000?
>>
>> Angus.
>>
>> > i'm lost
>> >
>> > On Mon, Jul 26, 2010 at 1:42 PM, <PH...@ge...> wrote:
>> >>
>> >> Make sure that you’re dividing by 1000.0 (as opposed to 1000) to avoid
>> >> any
>> >> integer nonsense and make sure to reset your axis limits.
>> >>
>> >>
>> >>
>> >> Posting a small code snippet would help us get you on the right track.
>> >>
>> >> -p
>> >>
>> >>
>> >>
>> >> From: Waléria Antunes David [mailto:wal...@gm...]
>> >> Sent: Monday, July 26, 2010 9:32 AM
>> >> To: Matthieu Brucher
>> >> Cc: mat...@li...
>> >> Subject: Re: [Matplotlib-users] Hz to KHz
>> >>
>> >>
>> >>
>> >> I know...I tried but I'm using django and also when divided by 1000 the
>> >> image does not appear...
>> >> I don't know what to do.....help me....
>> >>
>> >> On Mon, Jul 26, 2010 at 1:27 PM, Matthieu Brucher
>> >> <mat...@gm...> wrote:
>> >>
>> >> Hi,
>> >>
>> >> You may just divide them by 1000?
>> >>
>> >> Matthieu
>> >>
>> >> 2010/7/26 Waléria Antunes David <wal...@gm...>:
>> >>
>> >> > Hello all,
>> >> >
>> >> > I need to format the values of graphic to KHz.....my values are in Hz
>> >> > see at idle python it displays the values as: 3000 3050 3100 ....
>> >> > 3400 ,
>> >> > but
>> >> > I need to go where it will be displayed KHz:  3.0 3.1
>> >> >
>> >> > can someone help me?
>> >>
>> >> >
>> >> >
>> >> > ------------------------------------------------------------------------------
>> >> > The Palm PDK Hot Apps Program offers developers who use the
>> >> > Plug-In Development Kit to bring their C/C++ apps to Palm for a share
>> >> > of $1 Million in cash or HP Products. Visit us here for more details:
>> >> > http://ad.doubleclick.net/clk;226879339;13503038;l?
>> >> > http://clk.atdmt.com/CRS/go/247765532/direct/01/
>> >> > _______________________________________________
>> >> > Matplotlib-users mailing list
>> >> > Mat...@li...
>> >> > https://lists.sourceforge.net/lists/listinfo/matplotlib-users
>> >> >
>> >> >
>> >>
>> >>
>> >>
>> >> --
>> >> Information System Engineer, Ph.D.
>> >> Blog: http://matt.eifelle.com
>> >> LinkedIn: http://www.linkedin.com/in/matthieubrucher
>> >>
>> >>
>> >
>> >
>> > ------------------------------------------------------------------------------
>> > The Palm PDK Hot Apps Program offers developers who use the
>> > Plug-In Development Kit to bring their C/C++ apps to Palm for a share
>> > of $1 Million in cash or HP Products. Visit us here for more details:
>> > http://ad.doubleclick.net/clk;226879339;13503038;l?
>> > http://clk.atdmt.com/CRS/go/247765532/direct/01/
>> > _______________________________________________
>> > Matplotlib-users mailing list
>> > Mat...@li...
>> > https://lists.sourceforge.net/lists/listinfo/matplotlib-users
>> >
>> >
>>
>>
>>
>> --
>> AJC McMorland
>> Post-doctoral research fellow
>> Neurobiology, University of Pittsburgh
>
>



-- 
AJC McMorland
Post-doctoral research fellow
Neurobiology, University of Pittsburgh

Hi Everyone,

I am working on a weather conditions plot containing wind speed and
direction. For that I use a custom polar projection. Now updated from
Matplotlib 0.99 to 1.0 to use the subplot2grid feature.

Now, after trying and searching for hours, I seem to be too dumb to combine
my custom projection and subplot2grid. Before it did this way:

ax1=plt.figure().add_subplot(111, projection='northpolar')

But now I need sth like:

ax2 = plt.subplot2grid((5,5), (2,3), colspan=2, rowspan=3)

subplot2grid doesn't like a projection keyword.
How can I implement my projection? 

Thanks!
BB





-- 
View this message in context: http://old.nabble.com/matplotlib-subplot2grid-and-projection-tp29268717p29268717.html
Sent from the matplotlib - users mailing list archive at Nabble.com.

Replace this line:
  ax.plot(f/1000, Sserie)

With this line:
  ax.plot(f/1000.0, Sserie)

And tell us how things go. Python 2.6 distinguishes between integers and floats very strictly. Hence:
In [1]: 20/1000
Out[1]: 0

In [2]: 20.0/1000.0
Out[2]: 0.02

See the difference?
-paul

From: Waléria Antunes David [mailto:wal...@gm...]
Sent: Monday, July 26, 2010 10:01 AM
To: Angus McMorland
Cc: mat...@li...
Subject: Re: [Matplotlib-users] Hz to KHz

i don't understand..
On Mon, Jul 26, 2010 at 1:57 PM, Angus McMorland <am...@gm...<mailto:am...@gm...>> wrote:
On 26 July 2010 12:47, Waléria Antunes David <wal...@gm...<mailto:wal...@gm...>> wrote:
> My code like this in django: http://pastebin.com/nzM5jvuc
Are you perhaps suffering from integer division? How about dividing by
1000.0, instead of 1000?

Angus.

> i'm lost
>
> On Mon, Jul 26, 2010 at 1:42 PM, <PH...@ge...<mailto:PH...@ge...>> wrote:
>>
>> Make sure that you're dividing by 1000.0 (as opposed to 1000) to avoid any
>> integer nonsense and make sure to reset your axis limits.
>>
>>
>>
>> Posting a small code snippet would help us get you on the right track.
>>
>> -p
>>
>>
>>
>> From: Waléria Antunes David [mailto:wal...@gm...<mailto:wal...@gm...>]
>> Sent: Monday, July 26, 2010 9:32 AM
>> To: Matthieu Brucher
>> Cc: mat...@li...<mailto:mat...@li...>
>> Subject: Re: [Matplotlib-users] Hz to KHz
>>
>>
>>
>> I know...I tried but I'm using django and also when divided by 1000 the
>> image does not appear...
>> I don't know what to do.....help me....
>>
>> On Mon, Jul 26, 2010 at 1:27 PM, Matthieu Brucher
>> <mat...@gm...<mailto:mat...@gm...>> wrote:
>>
>> Hi,
>>
>> You may just divide them by 1000?
>>
>> Matthieu
>>
>> 2010/7/26 Waléria Antunes David <wal...@gm...<mailto:wal...@gm...>>:
>>
>> > Hello all,
>> >
>> > I need to format the values of graphic to KHz.....my values are in Hz
>> > see at idle python it displays the values as: 3000 3050 3100 .... 3400 ,
>> > but
>> > I need to go where it will be displayed KHz:  3.0 3.1
>> >
>> > can someone help me?
>>
>> >
>> > ------------------------------------------------------------------------------
>> > The Palm PDK Hot Apps Program offers developers who use the
>> > Plug-In Development Kit to bring their C/C++ apps to Palm for a share
>> > of $1 Million in cash or HP Products. Visit us here for more details:
>> > http://ad.doubleclick.net/clk;226879339;13503038;l?
>> > http://clk.atdmt.com/CRS/go/247765532/direct/01/
>> > _______________________________________________
>> > Matplotlib-users mailing list
>> > Mat...@li...<mailto:Mat...@li...>
>> > https://lists.sourceforge.net/lists/listinfo/matplotlib-users
>> >
>> >
>>
>>
>>
>> --
>> Information System Engineer, Ph.D.
>> Blog: http://matt.eifelle.com
>> LinkedIn: http://www.linkedin.com/in/matthieubrucher
>>
>>
>
> ------------------------------------------------------------------------------
> The Palm PDK Hot Apps Program offers developers who use the
> Plug-In Development Kit to bring their C/C++ apps to Palm for a share
> of $1 Million in cash or HP Products. Visit us here for more details:
> http://ad.doubleclick.net/clk;226879339;13503038;l?
> http://clk.atdmt.com/CRS/go/247765532/direct/01/
> _______________________________________________
> Matplotlib-users mailing list
> Mat...@li...<mailto:Mat...@li...>
> https://lists.sourceforge.net/lists/listinfo/matplotlib-users
>
>



--
AJC McMorland
Post-doctoral research fellow
Neurobiology, University of Pittsburgh

i don't understand..

On Mon, Jul 26, 2010 at 1:57 PM, Angus McMorland <am...@gm...> wrote:

> On 26 July 2010 12:47, Waléria Antunes David <wal...@gm...>
> wrote:
> > My code like this in django: http://pastebin.com/nzM5jvuc
>
> Are you perhaps suffering from integer division? How about dividing by
> 1000.0, instead of 1000?
>
> Angus.
>
> > i'm lost
> >
> > On Mon, Jul 26, 2010 at 1:42 PM, <PH...@ge...> wrote:
> >>
> >> Make sure that you’re dividing by 1000.0 (as opposed to 1000) to avoid
> any
> >> integer nonsense and make sure to reset your axis limits.
> >>
> >>
> >>
> >> Posting a small code snippet would help us get you on the right track.
> >>
> >> -p
> >>
> >>
> >>
> >> From: Waléria Antunes David [mailto:wal...@gm...]
> >> Sent: Monday, July 26, 2010 9:32 AM
> >> To: Matthieu Brucher
> >> Cc: mat...@li...
> >> Subject: Re: [Matplotlib-users] Hz to KHz
> >>
> >>
> >>
> >> I know...I tried but I'm using django and also when divided by 1000 the
> >> image does not appear...
> >> I don't know what to do.....help me....
> >>
> >> On Mon, Jul 26, 2010 at 1:27 PM, Matthieu Brucher
> >> <mat...@gm...> wrote:
> >>
> >> Hi,
> >>
> >> You may just divide them by 1000?
> >>
> >> Matthieu
> >>
> >> 2010/7/26 Waléria Antunes David <wal...@gm...>:
> >>
> >> > Hello all,
> >> >
> >> > I need to format the values of graphic to KHz.....my values are in Hz
> >> > see at idle python it displays the values as: 3000 3050 3100 .... 3400
> ,
> >> > but
> >> > I need to go where it will be displayed KHz:  3.0 3.1
> >> >
> >> > can someone help me?
> >>
> >> >
> >> >
> ------------------------------------------------------------------------------
> >> > The Palm PDK Hot Apps Program offers developers who use the
> >> > Plug-In Development Kit to bring their C/C++ apps to Palm for a share
> >> > of $1 Million in cash or HP Products. Visit us here for more details:
> >> > http://ad.doubleclick.net/clk;226879339;13503038;l?
> >> > http://clk.atdmt.com/CRS/go/247765532/direct/01/
> >> > _______________________________________________
> >> > Matplotlib-users mailing list
> >> > Mat...@li...
> >> > https://lists.sourceforge.net/lists/listinfo/matplotlib-users
> >> >
> >> >
> >>
> >>
> >>
> >> --
> >> Information System Engineer, Ph.D.
> >> Blog: http://matt.eifelle.com
> >> LinkedIn: http://www.linkedin.com/in/matthieubrucher
> >>
> >>
> >
> >
> ------------------------------------------------------------------------------
> > The Palm PDK Hot Apps Program offers developers who use the
> > Plug-In Development Kit to bring their C/C++ apps to Palm for a share
> > of $1 Million in cash or HP Products. Visit us here for more details:
> > http://ad.doubleclick.net/clk;226879339;13503038;l?
> > http://clk.atdmt.com/CRS/go/247765532/direct/01/
> > _______________________________________________
> > Matplotlib-users mailing list
> > Mat...@li...
> > https://lists.sourceforge.net/lists/listinfo/matplotlib-users
> >
> >
>
>
>
> --
> AJC McMorland
> Post-doctoral research fellow
> Neurobiology, University of Pittsburgh
>

On 26 July 2010 12:47, Waléria Antunes David <wal...@gm...> wrote:
> My code like this in django: http://pastebin.com/nzM5jvuc

Are you perhaps suffering from integer division? How about dividing by
1000.0, instead of 1000?

Angus.

> i'm lost
>
> On Mon, Jul 26, 2010 at 1:42 PM, <PH...@ge...> wrote:
>>
>> Make sure that you’re dividing by 1000.0 (as opposed to 1000) to avoid any
>> integer nonsense and make sure to reset your axis limits.
>>
>>
>>
>> Posting a small code snippet would help us get you on the right track.
>>
>> -p
>>
>>
>>
>> From: Waléria Antunes David [mailto:wal...@gm...]
>> Sent: Monday, July 26, 2010 9:32 AM
>> To: Matthieu Brucher
>> Cc: mat...@li...
>> Subject: Re: [Matplotlib-users] Hz to KHz
>>
>>
>>
>> I know...I tried but I'm using django and also when divided by 1000 the
>> image does not appear...
>> I don't know what to do.....help me....
>>
>> On Mon, Jul 26, 2010 at 1:27 PM, Matthieu Brucher
>> <mat...@gm...> wrote:
>>
>> Hi,
>>
>> You may just divide them by 1000?
>>
>> Matthieu
>>
>> 2010/7/26 Waléria Antunes David <wal...@gm...>:
>>
>> > Hello all,
>> >
>> > I need to format the values of graphic to KHz.....my values are in Hz
>> > see at idle python it displays the values as: 3000 3050 3100 .... 3400 ,
>> > but
>> > I need to go where it will be displayed KHz:  3.0 3.1
>> >
>> > can someone help me?
>>
>> >
>> > ------------------------------------------------------------------------------
>> > The Palm PDK Hot Apps Program offers developers who use the
>> > Plug-In Development Kit to bring their C/C++ apps to Palm for a share
>> > of $1 Million in cash or HP Products. Visit us here for more details:
>> > http://ad.doubleclick.net/clk;226879339;13503038;l?
>> > http://clk.atdmt.com/CRS/go/247765532/direct/01/
>> > _______________________________________________
>> > Matplotlib-users mailing list
>> > Mat...@li...
>> > https://lists.sourceforge.net/lists/listinfo/matplotlib-users
>> >
>> >
>>
>>
>>
>> --
>> Information System Engineer, Ph.D.
>> Blog: http://matt.eifelle.com
>> LinkedIn: http://www.linkedin.com/in/matthieubrucher
>>
>>
>
> ------------------------------------------------------------------------------
> The Palm PDK Hot Apps Program offers developers who use the
> Plug-In Development Kit to bring their C/C++ apps to Palm for a share
> of $1 Million in cash or HP Products. Visit us here for more details:
> http://ad.doubleclick.net/clk;226879339;13503038;l?
> http://clk.atdmt.com/CRS/go/247765532/direct/01/
> _______________________________________________
> Matplotlib-users mailing list
> Mat...@li...
> https://lists.sourceforge.net/lists/listinfo/matplotlib-users
>
>



-- 
AJC McMorland
Post-doctoral research fellow
Neurobiology, University of Pittsburgh

My code like this in django: http://pastebin.com/nzM5jvuc

i'm lost

On Mon, Jul 26, 2010 at 1:42 PM, <PH...@ge...> wrote:

>  Make sure that you’re dividing by 1000.0 (as opposed to 1000) to avoid
> any integer nonsense and make sure to reset your axis limits.
>
>
>
> Posting a small code snippet would help us get you on the right track.
>
> -p
>
>
>
> *From:* Waléria Antunes David [mailto:wal...@gm...]
> *Sent:* Monday, July 26, 2010 9:32 AM
> *To:* Matthieu Brucher
> *Cc:* mat...@li...
> *Subject:* Re: [Matplotlib-users] Hz to KHz
>
>
>
> I know...I tried but I'm using django and also when divided by 1000 the
> image does not appear...
> I don't know what to do.....help me....
>
> On Mon, Jul 26, 2010 at 1:27 PM, Matthieu Brucher <
> mat...@gm...> wrote:
>
> Hi,
>
> You may just divide them by 1000?
>
> Matthieu
>
> 2010/7/26 Waléria Antunes David <wal...@gm...>:
>
> > Hello all,
> >
> > I need to format the values of graphic to KHz.....my values are in Hz
> > see at idle python it displays the values as: 3000 3050 3100 .... 3400 ,
> but
> > I need to go where it will be displayed KHz:  3.0 3.1
> >
> > can someone help me?
>
> >
> ------------------------------------------------------------------------------
> > The Palm PDK Hot Apps Program offers developers who use the
> > Plug-In Development Kit to bring their C/C++ apps to Palm for a share
> > of $1 Million in cash or HP Products. Visit us here for more details:
> > http://ad.doubleclick.net/clk;226879339;13503038;l?
> > http://clk.atdmt.com/CRS/go/247765532/direct/01/
> > _______________________________________________
> > Matplotlib-users mailing list
> > Mat...@li...
> > https://lists.sourceforge.net/lists/listinfo/matplotlib-users
> >
> >
>
>
>
> --
> Information System Engineer, Ph.D.
> Blog: http://matt.eifelle.com
> LinkedIn: http://www.linkedin.com/in/matthieubrucher
>
>
>

It sounds like you're wanting a gaussian kernel density estimate (KDE) (not
the desktop!).  The other options you mentioned are for interpolation, and
are not at all what you're wanting to do.

You can use scipy.stats.kde.gaussian_kde()<http://www.scipy.org/doc/api_docs/SciPy.stats.kde.gaussian_kde.html>.
However, it currently doesn't take a weights array, so you'll need to modify
it for your use case.

If you prefer, I have faster version of a gaussian KDE that can take a
weights array.  It's actually slower than the scipy's gaussian kde for a low
number of points, but for hundreds, thousands, or millions of points, it's
several orders of magnitude faster.  (Though the speedup depends on the
covariance of the points... higher covariance = slower, generally speaking)

Here's a quick pastebin of the code.  http://pastebin.com/LNdYCZgw

To use it, you do something like the below... (assuming the code in the
pastebin is saved in a file called fast_kde.py)

import numpy as np
import matplotlib.pyplot as plt
from fast_kde import fast_kde

# From your description of your data...
weights, x, y = np.loadtxt('chain.txt', usecols=(0,4,6)).T

kde_grid = fast_kde(x, y, gridsize=(200,200), weights=weights)

# Plot the grid
plt.figure()
plt.imshow(kde_grid, extent=(x.min(), x.max(), y.max(), y.min())

# Reverse the y-axis
plt.gca().invert_yaxis()

plt.show()

Hope that helps a bit,
-Joe




On Sat, Jul 24, 2010 at 3:56 AM, montefra <fra...@go...>wrote:

>
> Hi,
>
> I am writing a program that reads three columns (one column containing the
> weights, the other two containing the values I want to plot) from a file
> containing the results from a MonteCarlo Markov Chain. The file contains
> thousends of lines. Then create the 2D histogram and make contourplots.
> Here
> is a sample of the code (I don't know if is correct, it's just to show what
> I do)
>
> >>> import numpy as np
> >>> import matplotlib.pyplot as mplp
> >>> chain = np.loadtxt("chain.txt", usecols=[0,4,6])   #read columns 0 (the
> >>> weights), 4 and 6 (the data), from the file "chain.txt"
> >>> h2D, xe, ye = np.histogram2D(chain[:,1],chain[:,2], weights=chain[:,0])
> >>> #create the 2D histogram
> >>> x = (xe[:-1] + xe[1:])/2. #x and y values for the plot (I use the mean
> >>> of each bin)
> >>> y = (ye[:-1] + ye[1:])/2.
> >>> mplp.figure()   #open the figure
> >>> mplp.contourf(x, y, h2D.T, origin='lower')  #contour plot
>
> As it is the contours are not smooth and they look not that nice. After
> days
> of searches I've found three methods and tried, unsuccesfully, to apply
> them
> 1) 2d interpolation: I got "segmentation fault" (on a quadcore machine with
> 8Gb of RAM)
> 2) Rbf (radial basis functions): I got wrong contours
> 3) ndimage: it creates spurious features (like secondary peaks parallel to
> the direction of the main one)
>
> Before beginning with Python, I used to use IDL to plot, and there is a
> function 'smooth' that smooth for you 2D histograms. I haven't found
> anything similar for Python.
> Does anyone have an idea or suggestion on how to do it?
>
> Thank in advance
> Francesco
>
> --
> View this message in context:
> http://old.nabble.com/Smooth-contourplots-tp29253884p29253884.html
> Sent from the matplotlib - users mailing list archive at Nabble.com.
>
>
>
> ------------------------------------------------------------------------------
> The Palm PDK Hot Apps Program offers developers who use the
> Plug-In Development Kit to bring their C/C++ apps to Palm for a share
> of $1 Million in cash or HP Products. Visit us here for more details:
> http://ad.doubleclick.net/clk;226879339;13503038;l?
> http://clk.atdmt.com/CRS/go/247765532/direct/01/
> _______________________________________________
> Matplotlib-users mailing list
> Mat...@li...
> https://lists.sourceforge.net/lists/listinfo/matplotlib-users
>

I know...I tried but I'm using django and also when divided by 1000 the
image does not appear...
I don't know what to do.....help me....

On Mon, Jul 26, 2010 at 1:27 PM, Matthieu Brucher <
mat...@gm...> wrote:

> Hi,
>
> You may just divide them by 1000?
>
> Matthieu
>
> 2010/7/26 Waléria Antunes David <wal...@gm...>:
> > Hello all,
> >
> > I need to format the values of graphic to KHz.....my values are in Hz
> > see at idle python it displays the values as: 3000 3050 3100 .... 3400 ,
> but
> > I need to go where it will be displayed KHz:  3.0 3.1
> >
> > can someone help me?
> >
> ------------------------------------------------------------------------------
> > The Palm PDK Hot Apps Program offers developers who use the
> > Plug-In Development Kit to bring their C/C++ apps to Palm for a share
> > of $1 Million in cash or HP Products. Visit us here for more details:
> > http://ad.doubleclick.net/clk;226879339;13503038;l?
> > http://clk.atdmt.com/CRS/go/247765532/direct/01/
> > _______________________________________________
> > Matplotlib-users mailing list
> > Mat...@li...
> > https://lists.sourceforge.net/lists/listinfo/matplotlib-users
> >
> >
>
>
>
> --
> Information System Engineer, Ph.D.
> Blog: http://matt.eifelle.com
> LinkedIn: http://www.linkedin.com/in/matthieubrucher
>