matplotlib-users — Discussion related to using matplotlib

Re: [Matplotlib-users] Re:Newbie Question...

[Ryugan M. <ryu...@sb...>]

I ran the following script on IDLE and the Boa-constructor.  I had a problem
on both of these when I
didn't include:

matplot.interactive(False)

Thanks for the info on matplotlibrc.

----- Original Message ----- 
From: "John Hunter" <jdh...@ac...>
To: "Ryugan Mizuta" <ryu...@sb...>
Cc: <mat...@li...>; "Barry Drake"
<bl...@ad...>
Sent: Wednesday, June 16, 2004 6:56 AM
Subject: Re: [Matplotlib-users] Re:Newbie Question...


> >>>>> "Ryugan" == Ryugan Mizuta <ryu...@sb...> writes:
>
>     Ryugan> Thanks for all who gave me a reply....  Somehow the
>     Ryugan> following script worked...:
>
>     import matplotlib
>     matplotlib.interactive(False)
>     matplotlib.use("WX")
>
>     from matplotlib.matlab import *
>     plot([1,2,3,4])
>     show()
>
>
> I just want to point out that in my opinion, the easiest way to set
> the default backend is to use your matplotlibrc file -
> http://matplotlib.sourceforge.net/faq.html#MATPLOTLIBRC.  For windows,
> this is found at C:\Python23\share\matplotlib\.matplotlibrc.
>
> You can achieve the same effect by setting
>
>     backend      : WX       # the default backend
>     interactive  : False    # see
http://matplotlib.sourceforge.net/interactive.html
>
> which will not be so invasive to your script (ie, you can change the
> default behavior of your scripts w/o changing the scripts
> themselves).  Using
>
>     import matplotlib
>     matplotlib.interactive(False)
>     matplotlib.use("WX")
>
> should be something you reserve to depart from the desired default
> behavior.
>
> Glad it's working for you!  I have two more questions I am curious
> about, for both you and Barry
>
>   * Does WXAgg also work?
>
>   * In which IDEs do you find this combination works?
>
> Following your question yesterday I added the FAQ entry
> http://matplotlib.sourceforge.net/faq.html#FREEZE, and I'd like to
> update it to make it as complete as possible.
>
> JDH

Re: [Matplotlib-users] Re:Newbie Question...

[Barry D. <bl...@ad...>]

John,
I ran the tests mentioned in a previous msg with both
WX and WXAgg.

Should also mention that the test with Idle doesn't
completely fail.  That is the graph window is
displayed and the program "appears" to terminate
normally.  It looks like some memory is being left
hanging and Idle, the OS, or matplotlib is trying to
write there on subsequent runs of the program (?).

Barry

--- John Hunter wrote:
> >>>>> "Barry" == Barry Drake 
> writes:
> 
>     Barry> Ryugan, Thanks.  That fixed my
problem as
> well: the graph
>     Barry> window disappearing after the call
to
> show().
> 
> Hi Barry,
> 
>     import matplotlib
>     matplotlib.interactive(False)
>     matplotlib.use("WX")
> 
>     from matplotlib.matlab import *
>     plot([1,2,3,4])
>     show()
> 
> Which line is responsible for the fix?  And are
you
> using WX or WXAgg?
> 
> JDH
> 
> 
>
-------------------------------------------------------
> This SF.Net email is sponsored by The 2004
> JavaOne(SM) Conference
> Learn from the experts at JavaOne(SM), Sun's
> Worldwide Java Developer
> Conference, June 28 - July 1 at the Moscone
Center
> in San Francisco, CA
> REGISTER AND SAVE! http://java.sun.com/javaone/sf
> Priority Code NWMGYKND
> _______________________________________________
> Matplotlib-users mailing list
> Mat...@li...
>
https://lists.sourceforge.net/lists/listinfo/matplotlib-users

Re: [Matplotlib-users] RE: how to freeze your matplotlib apps using py2exe

[John H. <jdh...@ac...>]

>>>>> "Colin" == Colin Rundel <ru...@ma...> writes:

    Colin> I'm trying to 'freeze' a script of mine using py2exe and I
    Colin> cant seem to get it to play nicely with matplotlib. More
    Colin> specifically, my script is a wxpython frame that displays
    Colin> several data plots that are created with the Agg backend
    Colin> and converted to bitmaps. From my limited knowledge of
    Colin> py2exe it appears that the problem arises from py2exe
    Colin> trying to import several backends at once.

Hi Colin, 

Sorry for the delay in getting back to you.  I just started playing
with py2exe.  I have an app that uses matplotlib that I would like to
freeze as well, so may as well take the plunge.

I found I could freeze a simple plot with 

    # setup.py
    from distutils.core import setup
    import glob
    import py2exe

    data = glob.glob(r'C:\Python23\share\matplotlib\*')
    data.append(r'C:\Python23\share\matplotlib\.matplotlibrc')

    setup(console=["simple_plot.py"],
          data_files=[("share",data)],
          )

where simple_plot.py is

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

I also got a similar warning message at the end like you about "The
following modules appear to be missing" but it appears to be harmless.

A simple_plot.exe was created in my dist dir, as were a number of dlls
and the matplotlib share data dir.  I had to set the MATPLOTLIB env
var to point to the share dir, ie

  set MATPLOTLIBDATA=c:\python\jdh\py2exe\simple_plot\dist\share

but then it (partly) worked.  TkAgg and WXAgg worked out of the box (I
could change the default backend in the rc file).  Plain vanilla WX
behaved strangely: The WX frame with navigation bars etc popped up but
no graph.  I could click 'save' and save the graph to a jpeg which
viewed correctly.  But no image in the wx frame.  Hmm.  Haven't gotten
GTK* working yet.

Are you using the latest py2exe?  Notably, I didn't get the gtk dll
error when I left out the excludes.

JDH

Re: [Matplotlib-users] Re:Newbie Question...

[Barry D. <bl...@ad...>]

John,
I tested with DrPython 2.4.2, PyPe 1.9, PythonWin,
Idle, SciTE 1.6, and, my usual dev environment,
UltraEdit 10.20b.  

Only SciTE hasn't worked so far; it exits with the
error:

Fatal Python error: PyEval_RestoreThread: NULL tstate

Even from a Win shell (DOS) the disappearing window
problem occurred until I inserted the line

matplotlib.interactive(False)

Idle seems to fail on the second run with error:

Unhandled exception at 0x1e0078d4 in pythonw.exe:
0xC0000005: Access violation reading location
0x00000004.

In the MS debugger I see the disassembled code:

1E0078CF  nop              
1E0078D0  mov         edx,dword ptr [esp+4] 
1E0078D4  mov         ecx,dword ptr [edx+4] 

where it breaks on the last line.

Thanks for the pointer on the matplotlibrc file.  I'll
switch to using that from now on (more *nix-like
anyway).

Barry

--- John Hunter wrote:
> >>>>> "Ryugan" == Ryugan Mizuta
> <ryu...@sb...> writes:
> 
>     Ryugan> Thanks for all who gave me a
reply.... 
> Somehow the
>     Ryugan> following script worked...:
> 
>     import matplotlib
>     matplotlib.interactive(False)
>     matplotlib.use("WX")
> 
>     from matplotlib.matlab import *
>     plot([1,2,3,4])
>     show()
> 
> 
> I just want to point out that in my opinion, the
> easiest way to set
> the default backend is to use your matplotlibrc
file
> -
>
http://matplotlib.sourceforge.net/faq.html#MATPLOTLIBRC.
>  For windows,
> this is found at
> C:\Python23\share\matplotlib\.matplotlibrc.
> 
> You can achieve the same effect by setting
> 
>     backend      : WX       # the default backend
>     interactive  : False    # see
>
http://matplotlib.sourceforge.net/interactive.html
> 
> which will not be so invasive to your script (ie,
> you can change the
> default behavior of your scripts w/o changing the
> scripts
> themselves).  Using
> 
>     import matplotlib
>     matplotlib.interactive(False)
>     matplotlib.use("WX")
> 
> should be something you reserve to depart from
the
> desired default
> behavior.
> 
> Glad it's working for you!  I have two more
> questions I am curious
> about, for both you and Barry
> 
>   * Does WXAgg also work?
> 
>   * In which IDEs do you find this combination
> works?
> 
> Following your question yesterday I added the FAQ
> entry
>
http://matplotlib.sourceforge.net/faq.html#FREEZE,
> and I'd like to
> update it to make it as complete as possible.
> 
> JDH
> 
> 
>
-------------------------------------------------------
> This SF.Net email is sponsored by The 2004
> JavaOne(SM) Conference
> Learn from the experts at JavaOne(SM), Sun's
> Worldwide Java Developer
> Conference, June 28 - July 1 at the Moscone
Center
> in San Francisco, CA
> REGISTER AND SAVE! http://java.sun.com/javaone/sf
> Priority Code NWMGYKND
> _______________________________________________
> Matplotlib-users mailing list
> Mat...@li...
>
https://lists.sourceforge.net/lists/listinfo/matplotlib-users

Re: [Matplotlib-users] Re:Newbie Question...

[John H. <jdh...@ac...>]

>>>>> "Barry" == Barry Drake <bl...@ad...> writes:

    Barry> Ryugan, Thanks.  That fixed my problem as well: the graph
    Barry> window disappearing after the call to show().

Hi Barry,

    import matplotlib
    matplotlib.interactive(False)
    matplotlib.use("WX")

    from matplotlib.matlab import *
    plot([1,2,3,4])
    show()

Which line is responsible for the fix?  And are you using WX or WXAgg?

JDH

Re: [Matplotlib-users] Re:Newbie Question...

[John H. <jdh...@ac...>]

>>>>> "Ryugan" == Ryugan Mizuta <ryu...@sb...> writes:

    Ryugan> Thanks for all who gave me a reply....  Somehow the
    Ryugan> following script worked...:

    import matplotlib
    matplotlib.interactive(False)
    matplotlib.use("WX")

    from matplotlib.matlab import *
    plot([1,2,3,4])
    show()


I just want to point out that in my opinion, the easiest way to set
the default backend is to use your matplotlibrc file -
http://matplotlib.sourceforge.net/faq.html#MATPLOTLIBRC.  For windows,
this is found at C:\Python23\share\matplotlib\.matplotlibrc.

You can achieve the same effect by setting

    backend      : WX       # the default backend
    interactive  : False    # see http://matplotlib.sourceforge.net/interactive.html

which will not be so invasive to your script (ie, you can change the
default behavior of your scripts w/o changing the scripts
themselves).  Using

    import matplotlib
    matplotlib.interactive(False)
    matplotlib.use("WX")

should be something you reserve to depart from the desired default
behavior.

Glad it's working for you!  I have two more questions I am curious
about, for both you and Barry

  * Does WXAgg also work?

  * In which IDEs do you find this combination works?

Following your question yesterday I added the FAQ entry
http://matplotlib.sourceforge.net/faq.html#FREEZE, and I'd like to
update it to make it as complete as possible.

JDH

Re: [Matplotlib-users] Re:Newbie Question...

[Barry D. <bl...@ad...>]

Ryugan,
Thanks.  That fixed my problem as well: the graph
window disappearing after the call to show().

Barry

--- Ryugan Mizuta wrote:
> Thanks for all who gave me a reply....
> 
> Somehow the following script worked...:
> 
> import matplotlib
> matplotlib.interactive(False)
> matplotlib.use("WX")
> 
> from matplotlib.matlab import *
> plot([1,2,3,4])
> show()
> 
> Thanks all for the help.
> 
> Ryugan
> 
> 

[Matplotlib-users] Re:Newbie Question...

[Ryugan M. <ryu...@sb...>]

Thanks for all who gave me a reply....

Somehow the following script worked...:

import matplotlib
matplotlib.interactive(False)
matplotlib.use("WX")

from matplotlib.matlab import *
plot([1,2,3,4])
show()

Thanks all for the help.

Ryugan

[Matplotlib-users] Re: two scale plots

[John H. <jdh...@ac...>]

>>>>> "Flavio" == Flavio Codeco Coelho <fcc...@fi...> writes:

    Flavio> how do I force the second plot to have the same limits on
    Flavio> the x axis as the first?

The short answer is you can't.  You could write some helper functions
or classes to force axes to behave similarly.  The example code below
will call any function on all the axes.  Use with caution! 

class CallAll:
    def __init__(self, seq, name):
        self.seq = seq
        self.name = name

    def __call__(self, *args, **kwargs):
        for a in self.seq:
            func = getattr(a, self.name)
            func(*args, **kwargs)

class SharedAxes:

   def __init__(self, axlist):
       self.axlist = axlist

   def __getattr__(self, key):
       return CallAll(self.axlist, key)

ax1 = subplot(211)
plot(blah, blah)

ax2 = subplot(212)
plot(blah, blah)

shared = SharedAxes((ax1, ax2))
shared.set_xlim(0,2)

Re: [Matplotlib-users] Newbie Question....

[Barry D. <bl...@ad...>]

Sorry, just realized that I didn't send this to the
list:

John,
I've been running the same examples  in all of the
IDE's with the WX backend without any problems.  I ran
them again when I saw this message.  To the examples I
just add the two lines:

import matplotlib
matplotlib.use("WXAgg")

and all is well.  Without these lines the graph
windows freeze (in WinXP).

However, after upgrading to version 0.5.4.2 there is a
new problem: after show() is run, the app quits and
the window disappears without any diagnostic output. 
This happens with all of the examples that I've tried
so far.  Before the upgrade they were working fine.

I haven't started debugging this yet.  Any ideas?

System:
Win XP Pro, 1GB RAM, Pentium 4 2.6GHz
Python 2.3.3, GTK-Runtime-Environment-2.2.4-3,
pygtk-2.2.0-1.win32-py2.3

Regaards,
Barry Drake
--- John Hunter wrote:
> >>>>> "Ryugan" == Ryugan Mizuta
> <ryu...@sb...> writes:
> 
>     Ryugan> Sorry, I'm a newbie and needed
some help
> to get started.
>     Ryugan> I downlaoded the Enthought
version(for
> WXPython) which is
>     Ryugan> recommended in the web-site and
> downloaded the latest
>     Ryugan> version of matplot.  I'm trying to
go
> through the tutorial
>     Ryugan> and I'm stuck with the first
example. 
> The following is
>     Ryugan> what I wrote down on the script:
> 
>     Ryugan> import matplotlib
matplotlib.use("WX")
> 
>     Ryugan> from matplotlib.matlab import *
> plot([1,2,3,4]) show()
> 
>     Ryugan> When I run this script it creates
a
> graph but the the
>     Ryugan> window freezes....
> 
>     Ryugan> Anybody have any suggestions on
how I
> can fix this...????
> 
> I'll bet dollars to donuts you are running
> matplotlib inside an IDE,
> eg Idle, PythonWin, Scintilla, etc.  Quoting from
> http://matplotlib.sourceforge.net/installing.html
> 
>   Important: There are known conflicts with some
of
> the backends with
>   some python IDEs such as pycrust, idle. If you
> want to use
>   matplotlib from an IDE, please consult the
> backends documentation
>   for compatibility information. You will have
the
> greatest likelihood
>   of success if you run the examples from the
> command shell or by
>   double clicking on them, rather than from an
IDE.
> If you are
>   interactively generating plots, your best bet
is
> TkAgg from the
>   standard python shell.
> 
> The "backends" page, referred to above, is at
> http://matplotlib.sourceforge.net/backends.html. 
> Basically, you
> should open up a command shell and try to run one
of
> the examples from
> the matplotlib src distribution (the *.zip file
for
> windows users);
> Eg,
> 
>   c:\matplotlib\examples> python
simple_plot.py
> 
> if this works fine (my guess is it will), then
> you'll know you have an
> IDE problem and not a matplotlib specific
problem. 
> Unfortunately,
> this problem tends to crop up a lot.  See for
> example
>
http://sourceforge.net/mailarchive/message.php?msg_id=8610351
> and the
> responses to that post for more information
> 
> 
> Let me know if this is the case. 
> 
> JDH
> 
> 
> 
>
-------------------------------------------------------
> This SF.Net email is sponsored by The 2004
> JavaOne(SM) Conference
> Learn from the experts at JavaOne(SM), Sun's
> Worldwide Java Developer
> Conference, June 28 - July 1 at the Moscone
Center
> in San Francisco, CA
> REGISTER AND SAVE! http://java.sun.com/javaone/sf
> Priority Code NWMGYKND
> _______________________________________________
> Matplotlib-users mailing list
> Mat...@li...
>
https://lists.sourceforge.net/lists/listinfo/matplotlib-users

Re: [Matplotlib-users] matplotlib usage: getting a python prompt at the end of a script

[John H. <jdh...@ac...>]

>>>>> "Istvan" == Istvan Albert <iu...@ps...> writes:

    Istvan> Hello Folks, At the end of a batch script using
    Istvan> matplotlib, once I close the plot window I'm getting a
    Istvan> python prompt:

    >>>>


This question comes up fairly often, so I added a FAQ entry to cover
it

  http://matplotlib.sourceforge.net/faq.html#PROMPT

Cheers!
JDH

[Matplotlib-users] matplotlib usage: getting a python prompt at the end of a script

[Istvan A. <iu...@ps...>]

Hello Folks,

At the end of a batch script using matplotlib, once I close the
plot window I'm getting a python prompt:

 >>>

that waits for input.

I searched the docs but I did not see this
mentioned. What do I need to do to have the
script exit once I close the plot window?

thanks,

Istvan.

Re: [Matplotlib-users] Newbie Question....

[John H. <jdh...@ac...>]

>>>>> "Ryugan" == Ryugan Mizuta <ryu...@sb...> writes:

    Ryugan> Sorry, I'm a newbie and needed some help to get started.
    Ryugan> I downlaoded the Enthought version(for WXPython) which is
    Ryugan> recommended in the web-site and downloaded the latest
    Ryugan> version of matplot.  I'm trying to go through the tutorial
    Ryugan> and I'm stuck with the first example.  The following is
    Ryugan> what I wrote down on the script:

    Ryugan> import matplotlib matplotlib.use("WX")

    Ryugan> from matplotlib.matlab import * plot([1,2,3,4]) show()

    Ryugan> When I run this script it creates a graph but the the
    Ryugan> window freezes....

    Ryugan> Anybody have any suggestions on how I can fix this...????

I'll bet dollars to donuts you are running matplotlib inside an IDE,
eg Idle, PythonWin, Scintilla, etc.  Quoting from
http://matplotlib.sourceforge.net/installing.html

  Important: There are known conflicts with some of the backends with
  some python IDEs such as pycrust, idle. If you want to use
  matplotlib from an IDE, please consult the backends documentation
  for compatibility information. You will have the greatest likelihood
  of success if you run the examples from the command shell or by
  double clicking on them, rather than from an IDE. If you are
  interactively generating plots, your best bet is TkAgg from the
  standard python shell.

The "backends" page, referred to above, is at
http://matplotlib.sourceforge.net/backends.html.  Basically, you
should open up a command shell and try to run one of the examples from
the matplotlib src distribution (the *.zip file for windows users);
Eg,

  c:\matplotlib\examples> python simple_plot.py

if this works fine (my guess is it will), then you'll know you have an
IDE problem and not a matplotlib specific problem.  Unfortunately,
this problem tends to crop up a lot.  See for example
http://sourceforge.net/mailarchive/message.php?msg_id=8610351 and the
responses to that post for more information


Let me know if this is the case. 

JDH

[Matplotlib-users] Newbie Question....

[Ryugan M. <ryu...@sb...>]

Sorry,  I'm a newbie and needed some help to get started.

I downlaoded the Enthought version(for WXPython)  which is recommended =
in the web-site and downloaded the latest version of matplot.  I'm =
trying to go through the tutorial and I'm stuck with the first example.  =
The following is what I wrote down on the script:

import matplotlib
matplotlib.use("WX")

from matplotlib.matlab import *
plot([1,2,3,4])
show()

When I run this script it creates a graph but the the window freezes....

Anybody have any suggestions on how I can fix this...????

Thanks.

Ryugan

Re: [Matplotlib-users] plot_date problem

[Peter G. <pgr...@ge...>]

Hi John:

> plot_date looks at the range of your date data and tries to pick the 
> appropriate
> date tick locator based on that range (ie a YearLocator, MonthLocator,
> MinuteLocator, etc). 


It does?! Do you mean that this is done automatically? Can you show me 
an example of this using only the time module (since I use python2.2, 
dont have datetime)? I thought that I had to manually set things up and 
tell matplotlib whether to use  YearLocator, MonthLocator, etc.. via 
calls to:   axes.xaxis.set_minor_locator(), 
axes.xaxis.set_major_locator(),  axes.xaxis.set_major_formatter().
In other words for every plot, check the time range of my data, figure 
out how many ticks I want, and decide whether to use months, days, 
hours, etc...
In fact, because I was getting some inconsistent results with using the 
above, I decided that for the most part (excluding a few special cases), 
I would print the time ticks myself 'manually'. The script  below shows 
what I mean.

On the other note, regarding the weird scaling that I talked about (and 
showed pretty pics for) in my last mail, I finally put together a small 
script that exposes the problem. It is a bit rough because I ripped bits 
and pieces from here and there, but shows the issue. Use the 
'wantBadPlot' and 'wantStandardDateTics' to see how things go wrong.

--------------------------------
#!/usr/bin/env python

import time
from matplotlib.dates import EpochConverter
from matplotlib.matlab import *
from matplotlib.ticker import FuncFormatter, NullLocator,  
MinuteLocator, DayLocator, HourLocator, MultipleLocator, DateFormatter

wantBadPlot=1
wantStandardDateTics=1
wantLegend=1

if wantBadPlot:
   time1=[1087192789.89]
   data1=[-65.54]
else:
   time1=[1087192289.89, 1087193789.89]
   data1=[-44.343, -65.54]
  time2=[
1087161589.89 ,
1087192289.0,
1087192389.0,
1087192489.0,
1087192589.0,
1087192689.0,
1087192789.89 ,
1087192889.0,
1087192989.0,
1087193089.0,
1087193189.0,
1087193289.0,
1087238100.0 ,
]
data2=[
-55.44
-64.54 ,
-66.54 ,
-61.54 ,
-69.54 ,
-45.66,
-55.54 ,
-77.54,
-65.54 ,
-49.54 ,
-57.54 ,
-68.54 ,
-55.54 ,
-23.44
]

ax = subplot(111)

p1Size=len(time1)
p2Size=len(time2)

p1=plot_date(time1, data1, None, '-', color='r')
p2=plot_date(time2, data2, None, '-', color='b')

if wantStandardDateTics:

   fmt=DateFormatter('%H:%M')
   hours=HourLocator(4)
     ax.xaxis.set_minor_locator(NullLocator())
   ax.xaxis.set_major_locator(hours)
   ax.xaxis.set_major_formatter(fmt)

   ax.autoscale_view()

else:
   #Manually display dates for tick-labels. Technically could use plot() 
and
   #get the same result.

   now=time2[-1]
   then=time2[0]
     deltaSec=now-then
   deltaTickSec=deltaSec/7.0

   tickList=[item for item in list(arange(then, now, deltaTickSec))]
     def tickString(x, pos):
       return time.strftime("%H:%M:%S", time.localtime(x))
     formatter = FuncFormatter(tickString)
     ax.set_xticks(tickList)
     ax.xaxis.set_major_formatter(formatter)
   ax.xaxis.set_minor_locator(NullLocator())
     ax.autoscale_view()

   #This will fix the problem!!
   #ax.set_xlim((then, now))

if wantLegend:
   legend((p1, p2), ('small data set (%d)' % p1Size, 'large data set  
(%d)' % p2Size))

xlabel('time')
grid(True)

show()
#savefig('./blah.png')
-------------------------------

Any ideas?

Finally, just want to verify (form my last email) that in axes.py:

 def get_ylim(self):
     "Get the y axis range [ymin, ymax]"
     return self.viewLim.intervalx().get_bounds()

should intervax() be intervaly()??

Thanks,
-- 

Peter Groszkowski         Gemini Observatory
Tel: +1 808 974-2509      670 N. A'ohoku Place
Fax: +1 808 935-9235      Hilo, Hawai'i 96720, USA




John Hunter wrote:

>>>>>> "Peter" == Peter Groszkowski <pio...@ho...> writes:
>>>>>>           
>>>>>
>
>    Peter> I found another issue with plot_date. Dont have a simple
>    Peter> example yet, and hope that this is something 'obvious' and
>    Peter> I don't have to bother.
>
> There is clearly something wrong with the autoscale function of one of
> the date tick locators.  It would help to know which one.  plot_date
> looks at the range of your date data and tries to pick the appropriate
> date tick locator based on that range (ie a YearLocator, MonthLocator,
> MinuteLocator, etc).  If I knew which tick locator was behaving badly,
> it would help me fix the problem.
>
> If you
>  print ax.xaxis._majorLocator
>
> after the call to plot_date, and let me know which locator it is, I
> can probably figure out where the problem is.  To simplify, don't
> explicitly set the date xlim range when you do this.
>
> On a side note, in your example code you call
>
>  ax.viewLim.intervalx().set_bounds(minXValueImPlotting, 
> maxXValueImPlotting)
>
> I assume you did this to narrow down the possible causes of problems.
> As you know, this is the call that ax.set_xlim makes under the hood.
> But in general, it's safest to stick to the axes API, ie, call
>
>  ax.set_xlim(minXValueImPlotting, maxXValueImPlotting)
>
> since this interface is guaranteed to be stable.
>
> JDH
>
>
> -------------------------------------------------------
> This SF.Net email is sponsored by the new InstallShield X.
> From Windows to Linux, servers to mobile, InstallShield X is the
> one installation-authoring solution that does it all. Learn more and
> evaluate today! http://www.installshield.com/Dev2Dev/0504
> _______________________________________________
> Matplotlib-users mailing list
> Mat...@li...
> https://lists.sourceforge.net/lists/listinfo/matplotlib-users
>  
>

Re: [Matplotlib-users] transparent canvas

[John H. <jdh...@ac...>]

>>>>> "Flavio" == Flavio Codeco Coelho <fcc...@fi...> writes:

    Flavio> Hi, This subject has already come up before and John
    Flavio> suggested using

    Flavio> f = figure(1, frameon=False)

    Flavio> for a totally transparent background which works fine as
    Flavio> an optional argumento to subplot in the matlab interface

    Flavio> subplot(111, frameon=False)
    Flavio> I want to know what would be the equivalent to

    Flavio> f._figurePatch.set_alpha(0.0)

    Flavio> for the matlab interface, since subplot does not have a
    Flavio> _figurePatch atribute.

I'm not sure exactly what you are asking.  Axes and Subplot don't have
a _figurePatch, but they do have an _axesPatch, which as you note you
can turn off by using frameon=False.  If you want to set it's alpha,
do

  ax = subplot(111)
  frame = ax.get_frame()
  frame.set_alpha(0.0)

Is this what you are looking for?

JDH

Re: [Matplotlib-users] axis label placement

[John H. <jdh...@ac...>]

>>>>> "Mark" == Mark Engelhardt <ma...@st...> writes:

    Mark> If I have axis labels with line returns in them, they end up
    Mark> overlapping with the plot area and other elements in the
    Mark> figure.  Is there a way to change their placement to move
    Mark> them away from the plot?  Is there a way to define the area
    Mark> occupied by a subplot to include the axis labels and titles
    Mark> such that multiple subplots do not have overlapping text?
    Mark> (This is related to another question I just posted, but I
    Mark> tried to separate these issues for easier referencing by
    Mark> others later on).

There is no way to do this automatically at present.  You can
explicitly define the size of your axes so that they don't overlap
each other, but this can take some trial and error.  subplot is just a
special case of axes that specifies the left, bottom, width, height
rectangle of the axes in a matlab consistent way.

The axes coordinates are normalized as fractions of the entire
figure.  See ganged_plots and axes_demo.py in the examples directory
of the matplotlib src distribution for examples of setting the axes
boundaries explicitly.

It is possible to write some automatic axes layout functions similar
to the ones you are looking for, eg, place axes 2 below axes 1 taking
into account tick labels and so on, but they doesn't currently exist.

It's also possible to write some GUI interaction code so these can be
dynamically resized - I can post some example code if you like.

The other things you may want to consider are

  * decreasing the tick padding (the spacing between the axes and tick label
    With the latest matplotlib this is controlled by tick.major.pad in
    the rc file

  * use a smaller tick font size, controlled by tick.labelsize

Hope this helps,
JDH

[Matplotlib-users] axis label placement

[Mark E. <ma...@st...>]

Hi,

I've spent a while searching for the answer to this, and I'm at the 
point of giving up... but it seems like it should be easy.

If I have axis labels with line returns in them, they end up 
overlapping with the plot area and other elements in the figure.  Is 
there a way to change their placement to move them away from the plot?  
Is there a way to define the area occupied by a subplot to include the 
axis labels and titles such that multiple subplots do not have 
overlapping text?  (This is related to another question I just posted, 
but I tried to separate these issues for easier referencing by others 
later on).

Thanks,
Mark

[Matplotlib-users] working with subplots

[Mark E. <ma...@st...>]

Hi,

Is there a way to change the spacing between subplots when using 
multiple subplots per figure?  I'm getting overlap between titles and 
x-labels.

Thanks,
Mark

Re: [Matplotlib-users] Vector Arrows in Image Plots

[Curtis C. <cu...@hi...>]

Mr. Hunter,

Thank you for your reply.  I am considering trying to develop the 2D
vector field plots myself using line collections, as you suggest.

Cheers,
Curtis


On Sat, 12 Jun 2004, John Hunter wrote:

> >>>>> "Curtis" == Curtis Cooper <cu...@hi...> writes:
>
>     Curtis> Hi, I am excited about the relatively new imshow and
>     Curtis> pcolor features in matplotlib.  I am using numarray to
>     Curtis> reduce/diagnose planetary atmospheres simulations, which
>     Curtis> consist of temperature and velocity data on gridpoints in
>     Curtis> longitude, latitude, and pressure.
>
>     Curtis> In addition to contour plots, which is on the goals list
>     Curtis> for this project, I would like a feature added that allows
>     Curtis> me to make velocity arrows overlaying the pcolor or imshow
>     Curtis> images, as is common in meteorology maps.  Will such a
>     Curtis> feature be available in matplotlib, and if so, how soon?
>     Curtis> Thanks for all the great work so far!
>
> Hi  Curtis,
>
> Actually this is on the goals page, listed as 2D vector plots - could
> be more descriptive there.  So yes, it is a priority.  But I can't
> really give an estimate at this point of when it will be done.
>
> Working in your favor is that it should be fairly easy now that line
> collections exist.  These were implemented in 0.54 and allow the fast
> drawing of a bunch of independent line segments, which both contours
> and vector fields have.  Now that the infrastructure exists to do
> these efficiently, it makes it more likely that they will be added
> sooner rather than later.  But there are a number of other equally important
> features on the TODO list, which is why I don't have a definite
> estimate.
>
> As the goals page states
>
>   Priorities can be changed in one of two ways: convincing one of the
>   existing developers to make an item a higher priority (for example,
>   if many users request the feature) or by helping to add the feature
>   yourself (i.e., becoming a new developer).
>
> You've just done option 1.  You're invited to try option 2!
>
> Cheers,
> JDH
>

[Matplotlib-users] Broken link

[Yann Le Du <yan...@no...>]

Hello,

Just to report a very minor thing, a broken link on the matplotlib website
: in the tutorial page, in the table summarising the "Lines properties",
the "color" hyperlink is broken.

-- 
Yann Le Du

http://www-thphys.physics.ox.ac.uk/users/YannLeDu

[Matplotlib-users] RE: how to freeze your matplotlib apps using py2exe

[Colin R. <ru...@ma...>]

I'm trying to 'freeze' a script of mine using py2exe and I cant seem to get 
it to play nicely with matplotlib. More specifically, my script is a 
wxpython frame that displays several data plots that are created with the 
Agg backend and converted to bitmaps. From my limited knowledge of py2exe 
it appears that the problem arises from py2exe trying to import several 
backends at once.

My specific setup is as follows:

Followed all the steps listed here: 
http://sourceforge.net/mailarchive/forum.php?thread_id=4031953&forum_id=33405
Also attempted using excludes as mentioned here: 
http://starship.python.net/crew/theller/moin.cgi/MatPlotLib

# setup.py
from distutils.core import setup
import py2exe
import glob

data = glob.glob(r"C:\Python23\share\matplotlib\*")

opts = {'py2exe': { 'excludes': ['_gtkagg', '_tkagg'],
                     	     'dll_excludes': ['libgdk-win32-2.0-0.dll', 
'libgobject-2.0-0.dll']
                          }
           }

setup(
     name="Sleepy",
     version="0.5.0",
     console=["sleepy.py"],
     data_files=[("share",data)],
     options = opt
)



#fsleepy.py import list if it helps at all
import wx
from wxPython.wx import *
from wxPython.gizmos import wxTreeListCtrl
from wxPython.lib.buttons import *

import matplotlib
import matplotlib.ft2font
import matplotlib.backends.backend_agg
import ttfquery
from matplotlib.matlab import *
matplotlib.use('Agg')

from matplotlib.backends.backend_agg import FigureCanvasAgg
from matplotlib.figure import Figure
from matplotlib.axes import Subplot

import win32com.client
import pickle, math, string, calendar, time, os, os.path



Finally, I'll attach output from two runs of py2exe on the target with and 
without the excludes listed in the py2exe wiki.

Thanks,
-Colin

Re: [Matplotlib-users] plot_date problem

[John H. <jdh...@ac...>]

>>>>> "Peter" == Peter Groszkowski <pio...@ho...> writes:

    Peter> I found another issue with plot_date. Dont have a simple
    Peter> example yet, and hope that this is something 'obvious' and
    Peter> I don't have to bother.

There is clearly something wrong with the autoscale function of one of
the date tick locators.  It would help to know which one.  plot_date
looks at the range of your date data and tries to pick the appropriate
date tick locator based on that range (ie a YearLocator, MonthLocator,
MinuteLocator, etc).  If I knew which tick locator was behaving badly,
it would help me fix the problem.

If you 

  print ax.xaxis._majorLocator

after the call to plot_date, and let me know which locator it is, I
can probably figure out where the problem is.  To simplify, don't
explicitly set the date xlim range when you do this.

On a side note, in your example code you call

  ax.viewLim.intervalx().set_bounds(minXValueImPlotting, maxXValueImPlotting)

I assume you did this to narrow down the possible causes of problems.
As you know, this is the call that ax.set_xlim makes under the hood.
But in general, it's safest to stick to the axes API, ie, call

  ax.set_xlim(minXValueImPlotting, maxXValueImPlotting)

since this interface is guaranteed to be stable.

JDH

Re: [Matplotlib-users] Vector Arrows in Image Plots

[John H. <jdh...@ac...>]

>>>>> "Curtis" == Curtis Cooper <cu...@hi...> writes:

    Curtis> Hi, I am excited about the relatively new imshow and
    Curtis> pcolor features in matplotlib.  I am using numarray to
    Curtis> reduce/diagnose planetary atmospheres simulations, which
    Curtis> consist of temperature and velocity data on gridpoints in
    Curtis> longitude, latitude, and pressure.

    Curtis> In addition to contour plots, which is on the goals list
    Curtis> for this project, I would like a feature added that allows
    Curtis> me to make velocity arrows overlaying the pcolor or imshow
    Curtis> images, as is common in meteorology maps.  Will such a
    Curtis> feature be available in matplotlib, and if so, how soon?
    Curtis> Thanks for all the great work so far!

Hi  Curtis,

Actually this is on the goals page, listed as 2D vector plots - could
be more descriptive there.  So yes, it is a priority.  But I can't
really give an estimate at this point of when it will be done.

Working in your favor is that it should be fairly easy now that line
collections exist.  These were implemented in 0.54 and allow the fast
drawing of a bunch of independent line segments, which both contours
and vector fields have.  Now that the infrastructure exists to do
these efficiently, it makes it more likely that they will be added
sooner rather than later.  But there are a number of other equally important
features on the TODO list, which is why I don't have a definite
estimate.

As the goals page states

  Priorities can be changed in one of two ways: convincing one of the
  existing developers to make an item a higher priority (for example,
  if many users request the feature) or by helping to add the feature
  yourself (i.e., becoming a new developer).

You've just done option 1.  You're invited to try option 2!

Cheers,
JDH

Re: [Matplotlib-users] plot_date problem

[Peter G. <pio...@ho...>]

hmm... forgot to attach the images...

---
I found another issue with plot_date. Dont have a simple example yet, and 
hope that this is something 'obvious' and I don't have to bother.

I attach two images showing the issue. I can go get rid of the weird scaling 
in 'bad.png'  when I do:

ax.viewLim.intervalx().set_bounds(minXValueImPlotting, maxXValueImPlotting)

after:

ax.autoscale_view()

I also noticed a couple of things inside axes.py, that might be wrong:

1)
   def get_ylim(self):
       "Get the y axis range [ymin, ymax]"
       return self.viewLim.intervalx().get_bounds()

should that intervax() be intervaly() ??

2) panx() and pany() are different as well, but perhaps they should be..

Changin these two things does not fix my problem so it has to be something 
else - hoping a simple typo. Will try to write a demo script that shows 
this... (I have matplotlib wrapped into other code so its not really a 
copy/paste-all type deal).

Peter


John Hunter wrote:

>>>>>>"Michael" == Michael Hauser <ha...@na...> writes:
>
>
>
>    Michael> Hello, I am having a problem with plot_date. I keep
>    Michael> getting the error:
>
>    Michael> Am I missing something obvious?
>
>No, there is a bug in plot_date in setting the tick formatter object.
>I didn't find this in any of my test or example scripts since all
>those explicitly set the formatter and hence hid the bug.  At the end
>of the Axes.plot_date function in matplotlib.axes, replace
>
>        self.xaxis.set_minor_locator(formatter)
>
>with        self.xaxis.set_major_formatter(formatter)
>
>and you'll be good to go.  Note however that there is a problem with
>your script in that the length of your x and y arrays must be the
>same.  After making the changes above, try
>
>from datetime import datetime
>from matplotlib.dates import EpochConverter
>from matplotlib.matlab import *
>
>times = [1084195314, 1084195375, 1084195436, 1084195497, 1084195557]
>vals = [10.2, 11.1, 8.7, 12.1, 12.2]
>converter = EpochConverter()
>ax = subplot(111)
>plot_date(times, vals, converter)
>savefig('test')
>show()
>
>
>
>
>-------------------------------------------------------
>This SF.Net email is sponsored by the new InstallShield X.
>From Windows to Linux, servers to mobile, InstallShield X is the
>one installation-authoring solution that does it all. Learn more and
>evaluate today! http://www.installshield.com/Dev2Dev/0504
>_______________________________________________
>Matplotlib-users mailing list
>Mat...@li...
>https://lists.sourceforge.net/lists/listinfo/matplotlib-users
>




-------------------------------------------------------
This SF.Net email is sponsored by the new InstallShield X.
From Windows to Linux, servers to mobile, InstallShield X is the
one installation-authoring solution that does it all. Learn more and
evaluate today! http://www.installshield.com/Dev2Dev/0504
_______________________________________________
Matplotlib-users mailing list
Mat...@li...
https://lists.sourceforge.net/lists/listinfo/matplotlib-users