matplotlib-users Mailing List for matplotlib (Page 6)

I have a program that uses matplotlib to plot two images.
figim24 = figure(figsize=(5,5))
figim70 = figure(figsize=(5,5))
I want an event loop that will be able to register which
of the two figures I mouse click in. When I only had one
figure I used
figim24.canvas.mpl_connect("button_press_event",clicker)
but that obviously only register events in figim24. Can anyone
guide me in the right direction for doing this?
Cheers
 Tommy

Hi,
Does anyone know how to create a multi-page postscript plot, so that 
several figures are included in one postscript file?
Thanks,
Rose
Rose Finn
Department of Physics
Siena College
Loudonville, NY 12211
(518) 782-6764

On Sunday 16 July 2006 19:38, Webb Sprague wrote:
> I have data with missing values represented by nans (like array([1.0,
> nan, 3.0]) that I am plotting with pylab.semilogy(). Unfortunately,
> the lines are getting connected through the nans, while I was hoping
> they would be left empty. If someone could tell me how to get empty
> lines, that would be great.
This is fixed in svn 2573.

On Sunday 16 July 2006 7:43 pm, PGM wrote:
> On Sunday 16 July 2006 19:38, Webb Sprague wrote:
> > I have data with missing values represented by nans (like array([1.0,
> > nan, 3.0]) that I am plotting with pylab.semilogy().
>
> Please transform your array in a MaskedArray.
No, this is a bug that should be fixed. I've submitted a bug report so I won't 
forget about it.

Masked arrays seem to do the trick.
Is there a reason why the nan thing won't work?
On 7/16/06, PGM <pgm...@gm...> wrote:
> On Sunday 16 July 2006 19:38, Webb Sprague wrote:
> > I have data with missing values represented by nans (like array([1.0,
> > nan, 3.0]) that I am plotting with pylab.semilogy().
>
> Please transform your array in a MaskedArray.
> import numpy as N
> masked_x=N.ma.masked_where(N.isnan(x),x)
>
> That should do the trick.
>
> Cf http://www.scipy.org/Cookbook/Matplotlib/Plotting_values_with_masked_arrays
>
>
> -------------------------------------------------------------------------
> Using Tomcat but need to do more? Need to support web services, security?
> Get stuff done quickly with pre-integrated technology to make your job easier
> Download IBM WebSphere Application Server v.1.0.1 based on Apache Geronimo
> http://sel.as-us.falkag.net/sel?cmd=lnk&kid=120709&bid=263057&dat=121642
> _______________________________________________
> Matplotlib-users mailing list
> Mat...@li...
> https://lists.sourceforge.net/lists/listinfo/matplotlib-users
>

On Sunday 16 July 2006 19:38, Webb Sprague wrote:
> I have data with missing values represented by nans (like array([1.0,
> nan, 3.0]) that I am plotting with pylab.semilogy().
Please transform your array in a MaskedArray.
import numpy as N
masked_x=N.ma.masked_where(N.isnan(x),x)
That should do the trick.
Cf http://www.scipy.org/Cookbook/Matplotlib/Plotting_values_with_masked_arrays

I have data with missing values represented by nans (like array([1.0,
nan, 3.0]) that I am plotting with pylab.semilogy(). Unfortunately,
the lines are getting connected through the nans, while I was hoping
they would be left empty. If someone could tell me how to get empty
lines, that would be great.
I get the behavior I expect with pylab.plot(), so if I need to I can
transform the data first, but I would rather get the log y-axis from
semilogy.
I am running version matplotlib 0.87.3 on Gentoo linux.
Thanks!

Works like a champ.
Thank you, S.
John Hunter wrote:
>>>>>> "Simon" == Simon Hook <sim...@jp...> writes:
>>>>>> 
>
> Simon> Hi, [cross posted to egenix and matplotlib]
>
> Simon> I have been using the egenix mxDateTime module and want to
> Simon> plot some of the dates with Matplotlib (pylab). Pylab uses
> Simon> num2date and date2num to covert datesandtimes to floating
> Simon> point values for plotting. Unfortunately if you give pylab
> Simon> date2num an egenix DateTime object, pylab bombs. Is there a
> Simon> simple way to use egenix DateTime objects with Pylab?
>
> See matplotlib.dates.mx2num
>
> JDH
> 

Simon Hook wrote:
> Hi,
> 
> [cross posted to egenix and matplotlib]
> 
> I have been using the egenix mxDateTime module and want to plot some of
> the dates with Matplotlib (pylab). Pylab uses num2date and date2num to
> covert datesandtimes to floating point values for plotting.
> Unfortunately if you give pylab date2num an egenix DateTime object,
> pylab bombs. Is there a simple way to use egenix DateTime objects with
> Pylab?
> 
> Below is a simple example
> 
> #D:\apps\Python23\python.exe
> 
> from datetime import *
> from mx.DateTime import *
> from pylab import *
> 
> # This works
> a=datetime(2005,10,10,5,5,5)
> print a
> 
> # This works
> b=DateTime(2005,10,10,5,5,5)
> print b
> 
> # This works
> c=date2num(a)
> print c
> 
> # This fails
> d=date2num(b)
> print d
> 
> The output and failure message is:
> 
> 2005年10月10日 05:05:05
> 2005年10月10日 05:05:05.00
> 732229.211863
I'm not sure what date2num() uses as epoch, but it looks like
some variant of a Julian Day Number:
>>> b.jdn
2453653.711863426
>>> b.tjd
3653.2118634259259
> Traceback (most recent call last):
> File "datetime-problem.py", line 20, in ?
> d=date2num(b)
> File "D:\apps\Python23\lib\site-packages\matplotlib\dates.py", line
> 174, in da
> te2num
> if not iterable(d): return _to_ordinalf(d)
> File "D:\apps\Python23\lib\site-packages\matplotlib\dates.py", line
> 137, in _t
> o_ordinalf
> base = dt.toordinal()
> AttributeError: toordinal
.toordinal() is a datetime method which is not supported by mxDateTime
instances.
-- 
Marc-Andre Lemburg
eGenix.com
Professional Python Services directly from the Source (#1, Jul 16 2006)
>>> Python/Zope Consulting and Support ... http://www.egenix.com/
>>> mxODBC.Zope.Database.Adapter ... http://zope.egenix.com/
>>> mxODBC, mxDateTime, mxTextTools ... http://python.egenix.com/
________________________________________________________________________
::: Try mxODBC.Zope.DA for Windows,Linux,Solaris,FreeBSD for free ! ::::

>>>>> "Simon" == Simon Hook <sim...@jp...> writes:
 Simon> Hi, [cross posted to egenix and matplotlib]
 Simon> I have been using the egenix mxDateTime module and want to
 Simon> plot some of the dates with Matplotlib (pylab). Pylab uses
 Simon> num2date and date2num to covert datesandtimes to floating
 Simon> point values for plotting. Unfortunately if you give pylab
 Simon> date2num an egenix DateTime object, pylab bombs. Is there a
 Simon> simple way to use egenix DateTime objects with Pylab?
See matplotlib.dates.mx2num
JDH

Hi,
[cross posted to egenix and matplotlib]
I have been using the egenix mxDateTime module and want to plot some of 
the dates with Matplotlib (pylab). Pylab uses num2date and date2num to 
covert datesandtimes to floating point values for plotting. 
Unfortunately if you give pylab date2num an egenix DateTime object, 
pylab bombs. Is there a simple way to use egenix DateTime objects with 
Pylab?
Below is a simple example
#D:\apps\Python23\python.exe
from datetime import *
from mx.DateTime import *
from pylab import *
# This works
a=datetime(2005,10,10,5,5,5)
print a
# This works
b=DateTime(2005,10,10,5,5,5)
print b
# This works
c=date2num(a)
print c
# This fails
d=date2num(b)
print d
The output and failure message is:
2005年10月10日 05:05:05
2005年10月10日 05:05:05.00
732229.211863
Traceback (most recent call last):
 File "datetime-problem.py", line 20, in ?
 d=date2num(b)
 File "D:\apps\Python23\lib\site-packages\matplotlib\dates.py", line 
174, in da
te2num
 if not iterable(d): return _to_ordinalf(d)
 File "D:\apps\Python23\lib\site-packages\matplotlib\dates.py", line 
137, in _t
o_ordinalf
 base = dt.toordinal()
AttributeError: toordinal
thanks, Simon

Muchas Muchas Gracias!
bpw
On 7/14/06, Stefan van der Walt <st...@su...> wrote:
> On Thu, Jul 13, 2006 at 08:21:14PM -0400, Darren Dale wrote:
> > On Thursday 13 July 2006 8:08 pm, Brian Wilfley wrote:
> > > I'm afraid I mixed and matched inappropriately withe the enthought 2.=
4
> > > beta 3 and matplotlibe 0.87.4 py2.4 pairing.
> > >
> > > Any thoughts?
> >
> > >
> > > RuntimeError: module compiled against version 90709 of C-API but this
> > > version of numpy is 90907
> >
> > I think the numpy version provided with enthought is pulled from the sv=
n
> > repository. If this is true, it will make life difficult for packages l=
ike
> > matplotlib for windows that are compiled against the latest numpy relea=
se, in
> > this case 0.9.8. Somebody please correct me if I'm wrong.
>
> >From http://code.enthought.com/enthon it looks like matplotlib is
> distributed along with the package, so there should be no need to use
> a separate install of matplotlib.
>
> Regards
> St=E9fan
>
>
> -------------------------------------------------------------------------
> Using Tomcat but need to do more? Need to support web services, security?
> Get stuff done quickly with pre-integrated technology to make your job ea=
sier
> Download IBM WebSphere Application Server v.1.0.1 based on Apache Geronim=
o
> http://sel.as-us.falkag.net/sel?cmd=3Dlnk&kid=3D120709&bid=3D263057&dat=
=3D121642
> _______________________________________________
> Matplotlib-users mailing list
> Mat...@li...
> https://lists.sourceforge.net/lists/listinfo/matplotlib-users
>
--=20
Brian P. Wilfley, Ph.D.
Director, Chief Scientist
Voice: 650.331.3476 x102
Fax: 650.887.2205
E-mail: bwi...@tr...
Triple Ring Technologies, Inc.
1850 Embarcadero Road
Palo Alto, CA 94303

I tried it out and it is fixed in the latest Enthought release 1.0.0beta4
Gary Ruben wrote:
> Hi Rob,
> 
> A couple of us reported this last week on the scipy list and I think it 
> should be fixed in the version which was just released by Enthought, so 
> if your friend will persevere and grab the latest version, it should be 
> OK - I hope to try it out today.
> 
> Gary R.
> 
> Rob Hetland wrote:
>> I am trying to help somebody get going on numpy/scipy/mpl. She is 
>> having trouble when starting enthon's matplotlib. She is a PC user (I 
>> am a unix/mac person), so I really don't know where to start finding the 
>> problem. Below is her note to me. Any advice would be helpful.
>>
>>
>> """
>> I got the enthought python etc. distribution (dated 7/5/06) again after
>> talking to you, and still get the same error with "from pylab import *"
>> Unfortunately, I can't grab the text from the ipython shell window, but
>> the gist is that pylab.m, tries to import Xaxis and Yaxis from axis,
>> axis.py tries to get FontProperties from font_manager.py, which in turn
>> looks for ft2font in matplotlib. Which is where the error message box
>> saying "entry point _ctype could not be located in the dll lib
>> msvcr71.dll" gets created. Maybe we should get another dummy tester to
>> try it. My system has several mscvr71.dll's and it's probably choosing to
>> use the wrong one....
>> """

>>>>> "Christopher" == Christopher Barker <Chr...@no...> writes:
 Christopher> well, duh. sorry for being such and idiot, but I'm
 Christopher> surprised that you can set numerix after importing
 Christopher> pylab.
You can't -- you need to change the numerix setting first (or use the
command line arg or the rc file)
import matplotlib
matplotlib.rcParams['numerix'] = 'numpy'
import pylab

Gary Ruben wrote:
> Yep, just do
> 
> from pylab import *
> rcParams['numerix'] = 'numpy'
well, duh. sorry for being such and idiot, but I'm surprised that you 
can set numerix after importing pylab.
> If I was developing something now, I would only bother supporting numpy.
That's my thought too, but I think I'll probably build Numeric an NumPy 
in anyway, as it's not hard to do.
Eric Firing wrote:
> For testing you may also take advantage of command-line arguments:
> 
> python examples/image_demo.py --Numeric
Excellent! that will make things much easier.
-Chris
-- 
Christopher Barker, Ph.D.
Oceanographer
 		
NOAA/OR&R/HAZMAT (206) 526-6959 voice
7600 Sand Point Way NE (206) 526-6329 fax
Seattle, WA 98115 (206) 526-6317 main reception
Chr...@no...

>>>>> "Richard" == Richard Albright <ral...@in...> writes:
 Richard> thanks for the subsy tip!, it did exactly what i couldn't
 Richard> do using NullLocator() and NullFormatter(), regardless if
 Richard> i put it before or after the semilogy command.
That's probably because you were setting the major locator and you
should have been setting the minor locator.
Eg,
 from pylab import subplot, semilogy, NullLocator, show
 x = 1,2,3
 y = 1,10,100
 ax = subplot(111)
 ax.semilogy(x, y)
 ax.yaxis.set_minor_locator(NullLocator())
 show()

thanks for the subsy tip!, it did exactly what i couldn't do using
NullLocator() and NullFormatter(), regardless if i put it before or
after the semilogy command.
On Fri, 2006年07月14日 at 11:52 -0500, John Hunter wrote:
> >>>>> "PGM" == PGM <pgm...@gm...> writes:
> 
> >> There is an easier way, however.
> 
> PGM> That's what I like in matplotlib: no matter how hard you try,
> PGM> there's always a simpler solution you're not yet aware of...
> 
> hmm... if you try hard, you should be led to the easy way! while
> subsx and subsy are documented in semilogx and semilogy, I added some
> detail to make this more obvious.
> 
> Cheers,
> JDH
> 
> 
> -------------------------------------------------------------------------
> Using Tomcat but need to do more? Need to support web services, security?
> Get stuff done quickly with pre-integrated technology to make your job easier
> Download IBM WebSphere Application Server v.1.0.1 based on Apache Geronimo
> http://sel.as-us.falkag.net/sel?cmd=lnk&kid=120709&bid=263057&dat=121642
> _______________________________________________
> Matplotlib-users mailing list
> Mat...@li...
> https://lists.sourceforge.net/lists/listinfo/matplotlib-users
-- 
Rick Albright
Senior Quantitvative Analyst
Indie Research, LLC
254 Witherspoon Street
Princeton, NJ 08542
(609)497-1030
ral...@in...

>>>>> "PGM" == PGM <pgm...@gm...> writes:
 >> There is an easier way, however.
 PGM> That's what I like in matplotlib: no matter how hard you try,
 PGM> there's always a simpler solution you're not yet aware of...
hmm... if you try hard, you should be led to the easy way! while
subsx and subsy are documented in semilogx and semilogy, I added some
detail to make this more obvious.
Cheers,
JDH

> There is an easier way, however. 
That's what I like in matplotlib: no matter how hard you try, there's always a 
simpler solution you're not yet aware of...

>>>>> "PGM" == PGM <pgm...@gm...> writes:
 PGM> On Friday 14 July 2006 11:25, Richard Albright wrote:
 >> from pylab import * x=(1,2,3,4,5) y=(13, 22,19,26,32)
 >> set_major_locator(NullLocator())
 >> set_major_formatter(NullFormatter()) semilogy(x,y) show()
 PGM> Have you tried that ?
 PGM> gca().yaxis.set_minor_locator(NullLocator()) Or maybe I don't
 PGM> understand what you're trying to do...
I'm also a little unclear as to what Richard wants to do, but if you
want to turn off the minor ticks, your approach is correct. You just
need to make sure you set the minor locator *after* the call to
semilogy.
There is an easier way, however. The kwargs subsy is a sequence of
locations where you want the minor ticks (eg 1,2,5) for base-10.
You can set this list to be empty to turn off all the minor ticks
 In [3]: semilogy(x,y,subsy=[])
JDH

On Friday 14 July 2006 11:25, Richard Albright wrote:
> from pylab import *
> x=(1,2,3,4,5)
> y=(13, 22,19,26,32)
> set_major_locator(NullLocator())
> set_major_formatter(NullFormatter())
> semilogy(x,y)
> show()
Have you tried that ?
gca().yaxis.set_minor_locator(NullLocator()) 
Or maybe I don't understand what you're trying to do...

How can I turn off the decade ticks in the following simple example?
from pylab import *
x=(1,2,3,4,5)
y=(13, 22,19,26,32)
set_major_locator(NullLocator())
set_major_formatter(NullFormatter())
semilogy(x,y)
show()
I am building stock graphs with the library, but have not been able to
figure out how to turn off the decade tick marks. If its possible, any
advice on how to do this would be greatly appreciated. 
-- 
Rick Albright
Senior Quantitvative Analyst
Indie Research, LLC
254 Witherspoon Street
Princeton, NJ 08542
(609)497-1030
ral...@in...

On Thu, Jul 13, 2006 at 08:21:14PM -0400, Darren Dale wrote:
> On Thursday 13 July 2006 8:08 pm, Brian Wilfley wrote:
> > I'm afraid I mixed and matched inappropriately withe the enthought 2.=
4
> > beta 3 and matplotlibe 0.87.4 py2.4 pairing.
> >
> > Any thoughts?
>=20
> >
> > RuntimeError: module compiled against version 90709 of C-API but this
> > version of numpy is 90907
>=20
> I think the numpy version provided with enthought is pulled from the sv=
n=20
> repository. If this is true, it will make life difficult for packages l=
ike=20
> matplotlib for windows that are compiled against the latest numpy relea=
se, in=20
> this case 0.9.8. Somebody please correct me if I'm wrong.
>From http://code.enthought.com/enthon it looks like matplotlib is
distributed along with the package, so there should be no need to use
a separate install of matplotlib.
Regards
St=E9fan

Till Wagner <sac...@ya...> writes:
> leg.draw_frame(False) might be an option, but I use
> self.axes.grid(True) and when the legend has no
> borders, the grid is shown through the text and makes
> ist nearly unredable.
I use something like this ("setp(legendframe, linewidth=3D0.0)" is the
crucial command):
# http://matplotlib.sourceforge.net/matplotlib.pylab.html#-legend
legend(loc=3D'upper left', shadow=3DFalse)
legend =3D gca().get_legend()
legendtext =3D legend.get_texts()
legendlines =3D legend.get_lines()
legendframe =3D legend.get_frame()
setp(legendtext, fontsize=3D'small')
setp(legendlines, linewidth=3D1.5)
#legend.draw_frame(False) # don't draw the legend frame
setp(legendframe, linewidth=3D0.0)
legendframe.set_facecolor(0.98) # set the frame face color to light =
gray
HTH,
 Colin

You are right, concerning your comment below.
That will work just fine,
Mark
On 7/13/06, Eric Firing <ef...@ha...> wrote:
>
>
>
> But why is this better than the following?
>
> plot(Z[0,:], Z[1:,:])
>
> The latter would accomplish the same, be completely consistent with
> option 4, be completely explicit and unambiguous, require no more typing
> than using a kwarg, require no extra logic in the plot code, and require
> no extra documentation for the plot command.
>
> Eric
> >
> > As you said, there will be many more opinions,
> >
> > Mark
> >
> >
> >
> > To summarize, the options seem to be:
> >
> > 1) Leave plot argument parsing alone.
> > 2) Accept an Nx2 array in place of a pair of arguments containing x
> > and y.
> >
> > 3) Implement the Matlab model.
> > 4) Implement the Matlab model, but taking rows instead of columns in
> an
> > X or Y array that is 2-D.
> >
> > I am open to arguments, but my preference is the Matlab model. I
> don't
> > think that the difference in native array storage order matters
> much.
> > It is more important to have the API at the plot method and function
> > level match the way people think.
> >
> > Eric
> >
> >
> >
> > ------------------------------------------------------------------------
> >
> >
> >
> -------------------------------------------------------------------------
> > Using Tomcat but need to do more? Need to support web services,
> security?
> > Get stuff done quickly with pre-integrated technology to make your job
> easier
> > Download IBM WebSphere Application Server v.1.0.1 based on Apache
> Geronimo
> > http://sel.as-us.falkag.net/sel?cmd=lnk&kid=120709&bid=263057&dat=121642
> >
> >
> > ------------------------------------------------------------------------
> >
> > _______________________________________________
> > Matplotlib-users mailing list
> > Mat...@li...
> > https://lists.sourceforge.net/lists/listinfo/matplotlib-users
>
>