matplotlib-users — Discussion related to using matplotlib

Re: [Matplotlib-users] EPS printing problems

[Stephen W. <ste...@cs...>]

Jochen Voss wrote:

>Last I checked the PostScript backend used only PostScript level 2.
>  
>
What I really meant was that the PostScript file produced by MPL starts 
with '%!PS-Adobe-3.0', which I thought meant Postscript level 3.  
Changing the 3.0 to 2.0 by hand still doesn't allow it to print.  dvips 
output files print fine on this printer, and they start with 
'%!PS-Adobe-2.0'.

>You could try to produce a standalone .ps file instead of an
>.eps file.  Maybe this works better?
>  
>
Thanks for the suggestion, but I'm afraid this didn't help.  The 
workaround I found was to do a dvipdf on the LaTeX file which included 
the MPL doc, read it with Acrobat Reader, and tell acroread to print it 
at Postscript level 1.  Normally acroread can print at level 2 to this 
printer, but asking acroread to print at level 3 gets an error page from 
the printer telling you to use level 2.

Best,
Steve

[Matplotlib-users] problem with multiaxis plot

[Wendell C. <wcr...@uf...>]

'''Hi,

I use the following program in a simulation class to demonstrate the
niftyness of matplotlib. The combination of Python, Numeric, and matplotlib
really makes a great set of tools for biologists who are largely
non-programmers these days. Thanks. 

I just did a big upgrade from 0.6-before pylab to 0.80. This is the only
tested program that didn't produce the same results before and after:

Attached are matlab and pylab png files.

Is there another way of doing this that gets back to the original output?

Also, is it just my imagination, or does 0.8 take longer to display graphs
than previous versions. I haven't done any formal timing, but the delay
seems longer.

Thanks,

Wendell Cropper '''


from pylab import *

r = 0.1
g = 0.002
h = 0.0002
m = 0.2

def derivs(x, t):

    d1dt = r* x[0] - g * x[0] * x[1]   # host

    d2dt = h * x[0] * x[1] - m * x[1]  #parasite

    return (d1dt, d2dt)

dt = 0.01  #delta t

tim  = arange(0.0, 125.0, dt)

x0 = (1500.0, 50.0)  #initial numbers of host, parasite

xint = rk4(derivs, x0, tim) #Runge-Kutta 4th order numerical integration

#print xint[0]
#print xint[1]

host, parasite = [[x[0] for x in xint], [x[1] for x in xint]]  #extract
variables

figure(1, [11.0, 11.0])  #size of window 

ax1 = subplot(211)
title('Host-Parasite using MatPlotLib')
L1 = plot(tim, host, 'b', linewidth=2)
plot ([0.0, 0.1], [500.0, 510.0], 'w') 
ax1.yaxis.tick_left()
ylabel('Host Numbers')

ax2 = subplot(211, frameon = False)
ax2.yaxis.tick_right()
text (145.0, 47.0, 'Parasite Numbers', rotation = 'vertical')
xlabel('time')
L2 = plot(tim, parasite, 'r--', linewidth=2)
plot ([0.0, 0.1], [90.0, 100.0], 'w')
legend((L1, L2), ('H', 'P'))



subplot(212)

plot(parasite, host, 'g')
axis([20.0, 100.0, 500.0, 1600.0]) #min x, max x, min y, max y
xlabel('Parasite Numbers')
ylabel('Host Numbers')
text (45.0, 1200.0, 'Phase Plane Plot', fontsize = 16)

show()

Re: [Matplotlib-users] Re: Memory usage

[Matt N. <new...@ca...>]

Hi Werner,

I think what you may want is something like this:

    def OnButton1Button(self, event):
        print 'done it'
        ### self.figure.clear()
        ### self.axes = self.figure.add_subplot(111)
        self.axes.cla()            # <-- clear the axes
        t = arange(0.0,4.0,0.01)
        s = sin(2*pi*t)

        self.axes.plot(t,s)
        self.axes.set_xlabel('Time 2 (s)')
        self.axes.set_ylabel('Price 2 ($)')
        self.canvas.draw()         # <-- force a redraw

Is that OK?  It works for me on Windows and Linux.  I do the
same thing for 'make a fresh plot' in my own codes that I know
work OK on Mac OS X as well.

Cheers,

--Matt

Re: [Matplotlib-users] Re: Memory usage

[Werner F. B. <wer...@fr...>]

#!/usr/bin/env python
"""
Show how to have wx draw a cursor over an axes that moves with the
mouse and reports the data coords
"""

from matplotlib.numerix import arange, sin, pi

import matplotlib
matplotlib.use('WXAgg')
from matplotlib.backends.backend_wxagg import FigureCanvasWxAgg as FigureCanvas
from matplotlib.backends.backend_wx import NavigationToolbar2Wx

from matplotlib.figure import Figure
import wx

class CanvasFrame(wx.Frame):

    def __init__(self):
        wx.Frame.__init__(self,None,-1,
                         'CanvasFrame',size=(550,350))

        self.SetBackgroundColour(wx.NamedColor("WHITE"))

        self.figure = Figure(frameon=False)
        self.axes = self.figure.add_subplot(111)
        t = arange(0.0,3.0,0.01)
        s = sin(2*pi*t)

        self.axes.plot(t,s)
        self.axes.set_xlabel('Time (s)')
        self.axes.set_ylabel('Price ($)')
        self.canvas = FigureCanvas(self, -1, self.figure)
        self.canvas.mpl_connect('motion_notify_event', self.mouse_move)
        self.sizer =wx.BoxSizer(wx.VERTICAL)
        self.sizer.Add(self.canvas, 1, wx.LEFT | wx.TOP | wx.GROW)
        buttonId = wx.NewId()
        self.button1 = wx.Button(id=buttonId, label='Clear and do figure again',
                        parent=self)
        self.button1.Bind(wx.EVT_BUTTON, self.OnButton1Button, id=buttonId)
        self.sizer.Add(self.button1, 0, wx.LEFT | wx.TOP)
        
        self.SetSizer(self.sizer)
        self.Fit()

        self.statusBar =wx.StatusBar(self, -1)
        self.statusBar.SetFieldsCount(1)
        self.SetStatusBar(self.statusBar)

        self.add_toolbar()  # comment this out for no toolbar

        if wx.VERSION_STRING >= '2.5':
            self.Bind(wx.EVT_PAINT, self.OnPaint)
        else:
            wx.EVT_PAINT(self, self.OnPaint)
            
        self.cnt = 0

    def OnButton1Button(self, event):
        if self.cnt == 0:
            self.figure.clear()
            self.axes = self.figure.add_subplot(111)
            t = arange(0.0,4.0,0.01)
            s = sin(2*pi*t*(self.cnt+1))
    
            self.axes.plot(t,s)
            self.axes.set_xlabel('Time 2 (s)')
            self.axes.set_ylabel('Price 2 ($)')
    
        if self.cnt == 2:
            self.figure.clear()
            self.axes = self.figure.add_subplot(111)
            t = arange(0.0,3.0,0.01)
            
            s = sin(2*pi*t*(self.cnt+1))    
            self.axes.plot(t,s)
            self.axes.set_xlabel('Time 3 (s)')
            self.axes.set_ylabel('Price 3 ($)')

        if self.cnt > 2:
            self.figure.clear()
            self.axes = self.figure.add_subplot(111)
            t = arange(0.0,2.0,0.01)
            
            s = sin(2*pi*t*(self.cnt+1))    
            self.axes.plot(t,s)
            self.axes.set_xlabel('Time 4 (s)')
            self.axes.set_ylabel('Price 4 ($)')
            
        self.cnt+=1
        self.canvas.Refresh()
        
    def mouse_move(self, event):
        self.draw_cursor(event)

    def add_toolbar(self):
        self.toolbar = NavigationToolbar2Wx(self.canvas)
        self.toolbar.Realize()
        tw, th = self.toolbar.GetSizeTuple()
        fw, fh = self.canvas.GetSizeTuple()
        self.toolbar.SetSize(wx.Size(fw, th))
        self.sizer.Add(self.toolbar, 0, wx.LEFT | wx.EXPAND)
        # update the axes menu on the toolbar
        self.toolbar.update()

    def OnPaint(self, event):
        self.erase_cursor()
        try: del self.lastInfo
        except AttributeError: pass
        self.canvas.draw()
        event.Skip()

    def draw_cursor(self, event):
        'event is a MplEvent.  Draw a cursor over the axes'
        if event.inaxes is None:
            self.erase_cursor()
            try: del self.lastInfo
            except AttributeError: pass
            return
        canvas = self.canvas
        figheight = canvas.figure.bbox.height()
        ax = event.inaxes
        left,bottom,width,height = ax.bbox.get_bounds()
        bottom = figheight-bottom
        top = bottom - height
        right = left + width
        x, y = event.x, event.y
        y = figheight-y

        dc =wx.ClientDC(canvas)
        dc.SetLogicalFunction(wx.XOR)
        wbrush =wx.Brush(wx.Colour(255,255,255), wx.TRANSPARENT)
        wpen =wx.Pen(wx.Colour(200, 200, 200), 1, wx.SOLID)
        dc.SetBrush(wbrush)
        dc.SetPen(wpen)

        dc.ResetBoundingBox()
        dc.BeginDrawing()

        x, y, left, right, bottom, top = [int(val) for val in x, y, left, right, bottom, top]

        self.erase_cursor()
        line1 = (x, bottom, x, top)
        line2 = (left, y, right, y)
        self.lastInfo = line1, line2, ax, dc
        dc.DrawLine(*line1) # draw new
        dc.DrawLine(*line2) # draw new
        dc.EndDrawing()

        time, price = event.xdata, event.ydata
        self.statusBar.SetStatusText("Time=%f  Price=%f"% (time, price), 0)

    def erase_cursor(self):
        try: lastline1, lastline2, lastax, lastdc = self.lastInfo
        except AttributeError: pass
        else:
            lastdc.DrawLine(*lastline1) # erase old
            lastdc.DrawLine(*lastline2) # erase old

class App(wx.App):

    def OnInit(self):
        'Create the main window and insert the custom frame'
        frame = CanvasFrame()
        frame.Show(True)

        return True

app = App(0)
app.MainLoop()

Re: [Matplotlib-users] Re: Memory usage

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

>>>>> "John" == John Hunter <jdh...@ac...> writes:

    John> matplotlib does cache some information, mainly text and font
    John> stuff) between draws which can cause memory to grow, but
    John> usually this is a small amount.  I'll explore further.

I wrote a little GUI neutral script to expose the same functionality.
This helps test whether the problem is backend specific or not.  If
you click anywhere in the axes, the sine wave updates.

I see the same effect: changing the sine wave frequency causes memory
usage to increase with each draw, but only for the first 50 or so
draws.  This appears to be an agg thing, since I see it on GTKAgg,
TkAgg, WXAgg but not GTK or WX.

However, I don't believe it is a memory leak.  Agg may be caching some
path or stroke information, because as I say the memory only seems to
increase when the path is changed.  However, after 50 or so draws, the
increase stops, so it does not appear to be a memory leak.  I often
see these kinds of effects when debugging memory leaks, and typically
discard the first 30-50 draws before assessing whether there is a leak
or not.

Here's the backend neutral script I am using, you can run it with
-dWXAgg, -dWX, -dGTKAgg etc to test various backends.  Note that I
also explicitly clear the text cache with Text.cached = {}

JDH

#!/usr/bin/env python
import os, sys, time, gc
from matplotlib.text import Text
from pylab import *

def report_memory(i):
    pid = os.getpid()
    a2 = os.popen('ps -p %d -o rss,sz' % pid).readlines()
    print i, '  ', a2[1],
    return int(a2[1].split()[1])


fig = figure()
ax = fig.add_subplot(111)
t = arange(0.0, 2.0, 0.01)
s = sin(2*pi*t)
ax.plot(t,s)

def click(event):
    fig.clear()
    s = sin(2*pi*t*(click.cnt+1))
    ax = fig.add_subplot(111)
    ax.plot(t,s)
    gc.collect()
    Text.cached = {}
    report_memory(click.cnt)
    fig.canvas.draw()
    click.cnt += 1
click.cnt = 1


    
#connect('motion_notify_event', report)
connect('button_press_event', click)
show()

Re: [Matplotlib-users] Re: Memory usage

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

>>>>> "Werner" == Werner F Bruhin <wer...@fr...> writes:

    Werner> Hi All, I put a small test case together based on
    Werner> wxcursor_demo.  When clicking the button1 it should show a
    Werner> new plot, but it does not.

It does for me...

Do you mean a new plot window, or just redraw in the current figure?
You may not see that it has redrawn the figure unless you somehow
change the plot (ed pan/zoom).  I modified your example so that each
time you redraw the frequency of the sine wave changes to make it more
apparent.

    Werner> I think my memory issue has to do with this as I created a
    Werner> new self.canvas each time, instead of doing a clear, but
    Werner> the big question is what do I need to call to show the new
    Werner> plot.

For one thing, add a gc.collect after you call clear to trigger the
garbage collector.

I added a memory report function and noticed something surprising.  If
I just plotted the same sine wave over and over, memory usage was
flat.  If I changed the sine wave frequency (still creating arrays of
the same size though) memory usage increased with each plot.

matplotlib does cache some information, mainly text and font stuff)
between draws which can cause memory to grow, but usually this is a
small amount.  I'll explore further.

Here is my lightly modified version of your test script.  The
report_memory function only works on unix like systems.

JDH

#!/usr/bin/env python
"""
Show how to have wx draw a cursor over an axes that moves with the
mouse and reports the data coords
"""



from matplotlib.numerix import arange, sin, pi

import matplotlib
matplotlib.use('WXAgg')
from matplotlib.backends.backend_wxagg import FigureCanvasWxAgg as FigureCanvas
from matplotlib.backends.backend_wx import NavigationToolbar2Wx

from matplotlib.figure import Figure
import wx


import os, sys, time, gc

def report_memory(i):
    pid = os.getpid()
    a2 = os.popen('ps -p %d -o rss,sz' % pid).readlines()
    print i, '  ', a2[1],
    return int(a2[1].split()[1])

class CanvasFrame(wx.Frame):

    def __init__(self):
        wx.Frame.__init__(self,None,-1,
                         'CanvasFrame',size=(550,350))

        self.SetBackgroundColour(wx.NamedColor("WHITE"))

        self.figure = Figure()
        self.axes = self.figure.add_subplot(111)
        t = arange(0.0,3.0,0.01)
        s = sin(2*pi*t)

        self.axes.plot(t,s)
        self.axes.set_xlabel('Time (s)')
        self.axes.set_ylabel('Price ($)')
        self.canvas = FigureCanvas(self, -1, self.figure)
        self.canvas.mpl_connect('motion_notify_event', self.mouse_move)
        self.sizer =wx.BoxSizer(wx.VERTICAL)
        self.sizer.Add(self.canvas, 1, wx.LEFT | wx.TOP | wx.GROW)
        buttonId = wx.NewId()
        self.button1 = wx.Button(id=buttonId, label='Clear and do figure again',
                        parent=self)
        self.button1.Bind(wx.EVT_BUTTON, self.OnButton1Button, id=buttonId)
        self.sizer.Add(self.button1, 0, wx.LEFT | wx.TOP)
        
        self.SetSizer(self.sizer)
        self.Fit()

        self.statusBar =wx.StatusBar(self, -1)
        self.statusBar.SetFieldsCount(1)
        self.SetStatusBar(self.statusBar)

        self.add_toolbar()  # comment this out for no toolbar

        if wx.VERSION_STRING >= '2.5':
            self.Bind(wx.EVT_PAINT, self.OnPaint)
        else:
            wx.EVT_PAINT(self, self.OnPaint)

        self.cnt = 0
    def OnButton1Button(self, event):
        print 'done it'
        self.figure.clear()
        self.axes = self.figure.add_subplot(111)
        t = arange(0.0,4.0,0.01)
        s = sin(2*pi*t*(self.cnt+1))
        #s = sin(2*pi*t)

        self.axes.plot(t,s)
        self.axes.set_xlabel('Time 2 (s)')
        self.axes.set_ylabel('Price 2 ($)')

        report_memory(self.cnt)
        self.cnt+=1
        gc.collect()
    def mouse_move(self, event):
        self.draw_cursor(event)

    def add_toolbar(self):
        self.toolbar = NavigationToolbar2Wx(self.canvas)
        self.toolbar.Realize()
        tw, th = self.toolbar.GetSizeTuple()
        fw, fh = self.canvas.GetSizeTuple()
        self.toolbar.SetSize(wx.Size(fw, th))
        self.sizer.Add(self.toolbar, 0, wx.LEFT | wx.EXPAND)
        # update the axes menu on the toolbar
        self.toolbar.update()

    def OnPaint(self, event):
        self.erase_cursor()
        try: del self.lastInfo
        except AttributeError: pass
        self.canvas.draw()
        event.Skip()

    def draw_cursor(self, event):
        'event is a MplEvent.  Draw a cursor over the axes'
        if event.inaxes is None:
            self.erase_cursor()
            try: del self.lastInfo
            except AttributeError: pass
            return
        canvas = self.canvas
        figheight = canvas.figure.bbox.height()
        ax = event.inaxes
        left,bottom,width,height = ax.bbox.get_bounds()
        bottom = figheight-bottom
        top = bottom - height
        right = left + width
        x, y = event.x, event.y
        y = figheight-y

        dc =wx.ClientDC(canvas)
        dc.SetLogicalFunction(wx.XOR)
        wbrush =wx.Brush(wx.Colour(255,255,255), wx.TRANSPARENT)
        wpen =wx.Pen(wx.Colour(200, 200, 200), 1, wx.SOLID)
        dc.SetBrush(wbrush)
        dc.SetPen(wpen)

        dc.ResetBoundingBox()
        dc.BeginDrawing()

        x, y, left, right, bottom, top = [int(val) for val in x, y, left, right, bottom, top]

        self.erase_cursor()
        line1 = (x, bottom, x, top)
        line2 = (left, y, right, y)
        self.lastInfo = line1, line2, ax, dc
        dc.DrawLine(*line1) # draw new
        dc.DrawLine(*line2) # draw new
        dc.EndDrawing()

        time, price = event.xdata, event.ydata
        self.statusBar.SetStatusText("Time=%f  Price=%f"% (time, price), 0)

    def erase_cursor(self):
        try: lastline1, lastline2, lastax, lastdc = self.lastInfo
        except AttributeError: pass
        else:
            lastdc.DrawLine(*lastline1) # erase old
            lastdc.DrawLine(*lastline2) # erase old

class App(wx.App):

    def OnInit(self):
        'Create the main window and insert the custom frame'
        frame = CanvasFrame()
        frame.Show(True)

        return True

app = App(0)
app.MainLoop()

[Matplotlib-users] Re: Character encoding - was I18n

[Werner F. B. <wer...@fr...>]

O.k., so I have to decode all the texts, so that they get passed as 
"u'text' to matplotlib, i.e. something along these lines:

     textp = (_('Purchase - %s') % y).decode('iso-8859-1')

Note that in this example the word 'Purchase' gets translated using 
gettext and would return "Einkäufe".

See you
Werner

Werner F. Bruhin wrote:
> Hi All,
> 
> The legend lables etc I pass to matplotlib contain things like "äü" 
> (i.e. European character set) but they don't get displayed in the plots.
> 
> Is there something special one needs to do to support this?
> 
> Using wxPython 2.5.5.1 ANSI build (can't switch to Unicode just yet) on 
> Windows (XP, 2000).
> 
> See you
> Werner
> 
> 
> 
> -------------------------------------------------------
> SF email is sponsored by - The IT Product Guide
> Read honest & candid reviews on hundreds of IT Products from real users.
> Discover which products truly live up to the hype. Start reading now.
> http://ads.osdn.com/?ad_id=6595&alloc_id=14396&op=click
> 

[Matplotlib-users] Re: Memory usage

[Werner F. B. <wer...@fr...>]

Hi All,


I put a small test case together based on wxcursor_demo.  When clicking 
the button1 it should show a new plot, but it does not.

I think my memory issue has to do with this as I created a new 
self.canvas each time, instead of doing a clear, but the big question is 
what do I need to call to show the new plot.

So, I am obviously missing something.

Any hints are appreciated.

See you
Werner

Werner F. Bruhin wrote:

> Hi All,
> 
> Doing multiple plots I see that memory usage grows and grows.
> 
> Before doing a new plot I do "self.figure.clear()", or 
> "self.figure.clf()", is there some other call I should use?
> 
> See you
> Werner
> 
> 
> 
> -------------------------------------------------------
> SF email is sponsored by - The IT Product Guide
> Read honest & candid reviews on hundreds of IT Products from real users.
> Discover which products truly live up to the hype. Start reading now.
> http://ads.osdn.com/?ad_id=6595&alloc_id=14396&op=click

Re: [Matplotlib-users] EPS printing problems

[Jochen V. <vo...@se...>]

Hello Stephen,

On Thu, Apr 14, 2005 at 04:16:49PM -0700, Stephen Walton wrote:
> I can't seem to print EPS files generated from matplotlib on my printer,=
=20
> a Minolta Magicolor 2200.  I think this is very specific to my setup but=
=20
> would be happy for suggestions.  My initial guess is that MPL is=20
> generating Postscript Level 3 and my printer only supports Level 2, but=
=20
> I'm open to other ideas.
Last I checked the PostScript backend used only PostScript level 2.
You could try to produce a standalone .ps file instead of an
=2Eeps file.  Maybe this works better?

All the best,
Jochen
--=20
http://seehuhn.de/

Re: [Matplotlib-users] bounding box in eps export -> change in release 0.8

[Darren D. <dd...@co...>]

Hi Sascha,

I can't think of anything that has changed that would cause this. But if you 
could be a little more specific about the conditions causing the behavior, 
I'll look into it. Posting a short script that reproduces the problem would 
be helpful.

Thanks,
Darren


On Friday 15 April 2005 5:32 am, Sascha Schnepp wrote:
> Hi,
>
> I realized a change in the behavior of the eps export with the release 0.8.
> Sometimes (not always!) the bounding box on the left side is very close to
> the ylabel. When converting this to a pdf and printing it, this results in
> a (partially or even totally) cutted ylabel on your printout...
>
> Greetings,
> Sascha
>
>
> -------------------------------------------------------
> SF email is sponsored by - The IT Product Guide
> Read honest & candid reviews on hundreds of IT Products from real users.
> Discover which products truly live up to the hype. Start reading now.
> http://ads.osdn.com/?ad_ide95&alloc_id396&op=Click
> _______________________________________________
> Matplotlib-users mailing list
> Mat...@li...
> https://lists.sourceforge.net/lists/listinfo/matplotlib-users

-- 
Darren S. Dale

Bard Hall
Department of Materials Science and Engineering
Cornell University
Ithaca, NY. 14850

dd...@co...

[Matplotlib-users] bounding box in eps export -> change in release 0.8

[Sascha S. <sc...@te...>]

Hi,

I realized a change in the behavior of the eps export with the release 0.=
8. Sometimes (not always!) the bounding box on the left side is very clos=
e to the ylabel. When converting this to a pdf and printing it, this resu=
lts in a (partially or even totally) cutted ylabel on your printout...

Greetings,
Sascha

Re: [Matplotlib-users] getting slices out of an array

[Olivier B. <Oli...@id...>]

Hello all,

> Yes, this is Numeric/numarray slicing, not list slicing.  When you get
> some time, you should read the pdf Numeric or numarray documentation.
> It's comprehensive and very good.

Thanks for all the answers. I'm now looking at the numarray/numeric
documentation as suggested.

Good day, and thanks again.

        Olivier
--=20
   . __    . ___  __.  | Olivier Bornet         Oli...@id...
  / /  `  / /  / /  /  | IDIAP             http://www.idiap.ch/~bornet/
 / /   / / /--/ /--'   | CP 592        http://www.idiap.ch/~bornet/pgp/
/ /__.' / /  / /       | CH-1920 Martigny           PGP-key: 0xC53D9218

Re: [Matplotlib-users] core dumped after importing matplotlib on SunOS

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

>>>>> "Olive" == Olive  <enc...@ya...> writes:

    Olive> Hello, I've juste downloaded the last version (0.8) of
    Olive> matplotlib and successfully built it on SunOS 5.8 with GTK
    Olive> backend (gcc 3.2.2).  "import matplolib" is OK but typing
    Olive> "from matplotlib.pylab import *" result in a core dumped !
    Olive> without any error message or exception...  "import gtk",
    Olive> "import pygtk", and "import wx" works OK.

    Olive> I've tried with version 0.72, same error :

I don't have a lot of good advice for you, but here is how I would go
about debugging the problem.  You can recompile mpl in verbose mode.
rm -rf "site-packages/matplotlib" and your "build" subdir and edit
setup.py and set VERBOSE=True and recompile/reinstall.

This can sometimes provide information about which extension code
function is failing.

You also want to create a test script that replicates the problem (in
your case it's a one-liner that simply imports matplotlib.pylab) and
run it with the --verbose-debug flag

   > python myscript.py --verbose-debug

This prints maximal debug info which can provide helpful information
to narrow down the possible culprits.

    Olive> Python 2.4.1 (#3, Apr 14 2005, 10:58:02) [GCC 3.2.2] on
    Olive> sunos5 Type "help", "copyright", "credits" or "license" for
    Olive> more information.
    >>>> from matplotlib.pylab import *
    Olive> Bus Error (core dumped)

The old-fashioned but effective way to triangulate this kind of
segfault is to put a lot of print statements into the module that is
segfaulting (in this case site-packages/matplotlib/pylab.py).  Space
these print statements every 50 lines or so

  print 'debug 1'
  ...50 lines of pylab...
  print 'debug 2'
  ...50 lines of pylab...
  print 'debug 3'

When you have narrowed the segfault to be between two numbers, say 2
and 3, add more statements between them, maybe with a delimiter such
as TAB '\t'

  print '\t debug 2a'
  ...5 lines of pylab...
  print '\t debug 2b'
  ...5 lines of pylab...
  print '\t debug 2c'
  ...5 lines of pylab...

and repeat as necessary with a finer grain until you find the actual
line causing the segfault.  Posting this info will go a long way to
actually discovering the bug.

I call this "poor man's debugging" or "manual binary search" and it's
primitive, ugly, and highly effective.

    Olive> One more thing is that after importing matplotlib, I got an
    Olive> error when using help :

Don't know if this is related or not...

JDH

Re: [Matplotlib-users] Auto locator for dates?

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

>>>>> "Joe" == Joe Wu <jw...@sd...> writes:

    Joe> Sorry if this question has been asked before: Has anyone
    Joe> written a locator class which acts like the AutoLocator but
    Joe> picks the best DateLocator based on current time span?

    Joe> For date plots, zooming in/out can yield no tick marks or too
    Joe> many tick marks, and I think using such a class can solve
    Joe> this problem.  Any suggestion for other approach?

Take a look at Axes.plot_date.  It inspects the date range and tries
to pick a good locator and formatter.  You could wrap this
functionality into an AutoDateLocator, following the model of
AutoLocator.

There is a slight wrinkle here in that you want an AutoDateFormatter
as well, but this shouldn't be any harder.  Again you can reuse the
logic of plot_date to pick the right format string dynamically.


If you succeed, please send it my way and I'll use it in plot_date.


JDH

[Matplotlib-users] core dumped after importing matplotlib on SunOS

[Olive <enc...@ya...>]

Hello,

I've juste downloaded the last version (0.8) of matplotlib and
successfully built it on SunOS 5.8 with GTK backend (gcc 3.2.2).
"import matplolib" is OK but typing "from matplotlib.pylab import *"
result in a core dumped ! without any error message or exception...
"import gtk", "import pygtk", and "import wx" works OK.

I've tried with version 0.72, same error :

Python 2.4.1 (#3, Apr 14 2005, 10:58:02)
[GCC 3.2.2] on sunos5
Type "help", "copyright", "credits" or "license" for more information.
>>> from matplotlib.pylab import *
Bus Error (core dumped)

One more thing is that after importing matplotlib, I got an error when
using help :

python
Python 2.4.1 (#3, Apr 14 2005, 10:58:02)
[GCC 3.2.2] on sunos5
Type "help", "copyright", "credits" or "license" for more information.
>>> import matplotlib
>>> help(matplotlib)
Traceback (most recent call last):
  File "<stdin>", line 1, in ?
  File "/opt/SunOS/5.8/lib/python2.4/site.py", line 328,
in__call__
    return pydoc.help(*args, **kwds)
  File "/opt/SunOS/5.8/lib/python2.4/pydoc.py", line
1647, in __call__
    self.help(request)
  File "/opt/SunOS/5.8/lib/python2.4/pydoc.py", line
1691, in help
    else: doc(request, 'Help on %s:')
  File "/opt/SunOS/5.8/lib/python2.4/pydoc.py", line
1475, in doc
    pager(title % desc + '\n\n' + text.document(object, name))
  File "/opt/SunOS/5.8/lib/python2.4/pydoc.py", line
295, in document
    if inspect.ismodule(object): return self.docmodule(*args)
  File "/opt/SunOS/5.8/lib/python2.4/pydoc.py", line
1069, in docmodule
    contents.append(self.document(value, key, name))
  File "/opt/SunOS/5.8/lib/python2.4/pydoc.py", line
296, in document
    if inspect.isclass(object): return self.docclass(*args)
  File "/opt/SunOS/5.8/lib/python2.4/pydoc.py", line
1193, in docclass
    lambda t: t[1] == 'method')
  File "/opt/SunOS/5.8/lib/python2.4/pydoc.py", line
1143, in spill
    name, mod, object))
  File "/opt/SunOS/5.8/lib/python2.4/pydoc.py", line
297, in document
    if inspect.isroutine(object): return self.docroutine(*args)
  File "/opt/SunOS/5.8/lib/python2.4/pydoc.py", line
1254, in docroutine
    doc = getdoc(object) or ''
  File "/opt/SunOS/5.8/lib/python2.4/pydoc.py", line 75,
in getdoc
    result = inspect.getdoc(object) or inspect.getcomments(object)
  File "/opt/SunOS/5.8/lib/python2.4/inspect.py", line
448, in getcomments
    lines, lnum = findsource(object)
  File "/opt/SunOS/5.8/lib/python2.4/inspect.py", line
437, in findsource
    if pat.match(lines[lnum]): break
IndexError: list index out of range

I really don't know where to begin to isolate the problem.
Do I missed an option in the build process to have some kind of debug ?
What can I do to help you better understand the problem ?

Thank you in advance for your help,

[Matplotlib-users] Auto locator for dates?

[Joe W. <jw...@sd...>]

Sorry if this question has been asked before: Has anyone written a locator
class which acts like the AutoLocator but picks the best DateLocator based
on current time span?

For date plots, zooming in/out can yield no tick marks or too many tick
marks, and I think using such a class can solve this problem.  Any
suggestion for other approach?

Thanks,

- Joe Wu

[Matplotlib-users] EPS printing problems

[Stephen W. <ste...@cs...>]

Hi, all,

I can't seem to print EPS files generated from matplotlib on my printer, 
a Minolta Magicolor 2200.  I think this is very specific to my setup but 
would be happy for suggestions.  My initial guess is that MPL is 
generating Postscript Level 3 and my printer only supports Level 2, but 
I'm open to other ideas.

Steve Walton

Re: [Matplotlib-users] How to get ticks on top of Polygons?

[Michael B. <mb...@jp...>]

On Wed, 13 Apr 2005, John Hunter wrote:

> One possible solution is to move the axis
> drawing commands to the end of the Axes.draw function.  Off the top of
> my head, I don't see any problem with this approach.  Typically, you
> want the ticks visible.

I agree.  I was thinking that the ticks would have to be dumped into the 
'artists' list and drawn in zorder, but it's much easier to just move the 
x/yaxis draw() calls after the 'artists' draw() calls.

Attached is a patch that does that.  I ran the examples/backend_driver.py 
with it in place and didn't see any problems.

--Michael

========================================================================
  Michael Brady                                      Phone: 818-354-4957
  Jet Propulsion Laboratory (M/S 301-140L)             Fax: 818-393-6388
  4800 Oak Grove Drive
  Pasadena, CA  91109                 E-mail: Mic...@jp... 
========================================================================

[Matplotlib-users] Cartographic plotting using matplotlib

[Sean G. <sgi...@fr...>]

Hi all,

I'm the author of a cartographic library for Python and have been 
exploring the use of matplotlib and Agg instead of MapServer/GD for 
rendering map features.  It was matplotlib's Basemap module that first 
caught my eye, and I promised its author that I would keep him up to 
date with my experiments.  For the curious, results are here:

   http://zcologia.com/news/25

I'm really quite keen on Agg now.  Blows GD away for stylishness and 
seems like it will be fast enough to use in high performance web maps.

While I am primarily interested in programming web map services, it 
occurs to me that some of the same functionality could be useful to 
users in a pylab environment.  Plotting features from GIS datasets, 
consuming web map and feature services, etc.  If anyone is considering 
this kind of effort, please let me know, I'd be happy to pitch in.

cheers,
Sean

--
Sean Gillies
sgillies at frii dot com
http://users.frii.com/sgillies

Re: [Matplotlib-users] getting slices out of an array

[Chris B. <Chr...@no...>]

Olivier Bornet wrote:

> This is exactly what I'm searching for. But this is not really a Python
> syntax... The coma in the square brackets is not standard for lists.
> Maybe this come from the array type ?

Yes, it does. I you are doing anything MATLAB-y with Python, you really 
want to know about either numarray or Numeric. numarray is a little bot 
more like MATLAB (it allows array indexing), and Numeric has better 
performance with small arrays, but either will do for most uses. It 
looks like we're on the way to a grand unification of the two anyway.

Make sure you go find the docs f0or the package you choose, and read 
through them, I think you'll like them a lot!

By the way: aside from the looping, when you do:

x = A[i][j]

rather than:

x = A[i,j]

there is a performance hit because the first version is creating a new 
array out of A[i], then indexing into that, rather than just indexing 
directly into the 2-d (or more d) array. This is particularly pronounced 
with numarray, as the array creation overhead is larger than with Numeric.

-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...

Re: [Matplotlib-users] getting slices out of an array

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

>>>>> "Olivier" == Olivier Bornet <Oli...@id...> writes:

    Olivier> Cool. :-D This is exactly what I'm searching for. But
    Olivier> this is not really a Python syntax... The coma in the
    Olivier> square brackets is not standard for lists.  Maybe this
    Olivier> come from the array type ?

Yes, this is Numeric/numarray slicing, not list slicing.  When you get
some time, you should read the pdf Numeric or numarray documentation.
It's comprehensive and very good.

Numarray also offers enhanced indexing ( Numeric does not yet) which
is a convenience you probably expect coming from matlab.  Eg, you can
index an array with a sequence of integers (you must be careful that
the sequence is not a tuple however; lists or arrays of ints are ok)

    >>> import numarray as na
    >>> a = na.arange(0.0, 2.0, 0.1)
    >>> ind = [2,5,7]
    >>> a[ind]
    array([ 0.2,  0.5,  0.7])

JDH

Re: [Matplotlib-users] markeredgecolor

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

>>>>> "Sascha" == Sascha Schnepp <sc...@te...> writes:

    Sascha> I often don't know which attributes can be used as
    Sascha> kwargs...    

Any Artist method that starts with 'set_' can be set with a kwarg.
Eg, if set_markeredgecolor is a method, you can do
markeredgecolor=something as a kwarg.

Read through the classdocs at
http://matplotlib.sourceforge.net/classdocs.html for the object of
interest, eg for lines (Line2D), see
http://matplotlib.sourceforge.net/matplotlib.lines.html

Also, you can call set(object) on any object and it will return the
current properties and their values (see also example/set_and_get.py)

BTW, the set and get introspection functionality was recently moved out of 
the pylab interface and now resides in matplotlib.artist so API
developers can use it as well

In [1]: lines = plot([1,2,3])
 
In [2]: set(lines)
    alpha: float
    antialiased or aa: [True | False]
    clip_box: a matplotlib.transform.Bbox instance
    clip_on: [True | False]
    color or c: any matplotlib color - see help(colors)
    dashes: sequence of on/off ink in points
    data: (array xdata, array ydata)
    data_clipping: [True | False]
    figure: a matplotlib.figure.Figure instance
    label: any string
    linestyle or ls: [ '-' | '--' | '-.' | ':' | 'steps' | 'None' ]
    linewidth or lw: float value in points
    lod: [True | False]
    marker: [ '+' | ',' | '.' | '1' | '2' | '3' | '4' | '<' | '>' | 'D' | 'H' | '^' | '_' | 'd' | 'h' | 'o' | 'p' | 's' | 'v' | 'x' | '|' ]
    markeredgecolor or mec: any matplotlib color - see help(colors)
    markeredgewidth or mew: float value in points
    markerfacecolor or mfc: any matplotlib color - see help(colors)
    markersize or ms: float
    transform: a matplotlib.transform transformation instance
    visible: [True | False]
    xclip: (xmin, xmax)
    xdata: array
    yclip: (ymin, ymax)
    ydata: array
    zorder: any number
 
In [3]:

Re: [Matplotlib-users] getting slices out of an array

[John G. <jn...@eu...>]

Lists are 1-dimensional, so the comma doesn't really make sense.

You can, however, slice lists in python eg

 >>> x = range(100)
 >>> x[10:100:5]
[10, 15, 20, 25, 30, 35, 40, 45, 50, 55, 60, 65, 70, 75, 80, 85, 90, 95]

You're right, that the commas come into play when you are dealing with 
the numarray (or numeric) arrays, where you may well have more than one 
dimension.    It is a natural extension of the python slice syntax to 
multiple dimensions.

John


Olivier Bornet wrote:

>Hi,
>
>On Fri, Apr 15, 2005 at 12:32:45AM +1000, Tim Leslie wrote:
>  
>
>>You'll be pleased to know that the slicing works in almost exactly the
>>same in way python. Try:
>>
>>x2 = x[1:4,0:4]
>>    
>>
>
>Cool. :-D
>This is exactly what I'm searching for. But this is not really a Python
>syntax... The coma in the square brackets is not standard for lists.
>Maybe this come from the array type ?
>
>  
>
>>The indexing is slightly different, but I'm sure you can work it out.
>>    
>>
>
>Yes, no problem.
>
>Thanks for your help.
>
>        Olivier
>  
>

Re: [Matplotlib-users] getting slices out of an array

[Olivier B. <Oli...@id...>]

Hi,

On Fri, Apr 15, 2005 at 12:32:45AM +1000, Tim Leslie wrote:
> You'll be pleased to know that the slicing works in almost exactly the
> same in way python. Try:
>=20
> x2 =3D x[1:4,0:4]

Cool. :-D
This is exactly what I'm searching for. But this is not really a Python
syntax... The coma in the square brackets is not standard for lists.
Maybe this come from the array type ?

> The indexing is slightly different, but I'm sure you can work it out.

Yes, no problem.

Thanks for your help.

        Olivier
--=20
   . __    . ___  __.  | Olivier Bornet         Oli...@id...
  / /  `  / /  / /  /  | IDIAP             http://www.idiap.ch/~bornet/
 / /   / / /--/ /--'   | CP 592        http://www.idiap.ch/~bornet/pgp/
/ /__.' / /  / /       | CH-1920 Martigny           PGP-key: 0xC53D9218

Re: [Matplotlib-users] getting slices out of an array

[Tim L. <ti...@cs...>]

On Thu, 14 Apr 2005, Olivier Bornet <Oli...@id...> wrote...

> Hello all,
> 
> I'm converting some matlab scripts to matlibplot, and I don't know how
> to make some "slicing" efficently with matlibplot. What I want to do is
> to translate a matlab code like:

You'll be pleased to know that the slicing works in almost exactly the
same in way python. Try:

x2 = x[1:4,0:4]

The indexing is slightly different, but I'm sure you can work it out.
 
Tim

> 
>     x = [[01, 02, 03, 04, 05, 06];
>          [11, 12, 13, 14, 15, 16];
>          [21, 22, 23, 24, 25, 26];
>          [31, 32, 33, 34, 35, 36];
>          [41, 42, 43, 44, 45, 46];
>          [51, 52, 53, 54, 55, 56]
>          ];
>     x(2:4,1:4)
>     
>     ans =
>     
>         11    12    13    14
>         21    22    23    24
>         31    32    33    34
>     
> What I have done now in matplotlib is like:
> 
>     x = array ([[01, 02, 03, 04, 05, 06],
>                 [11, 12, 13, 14, 15, 16],
>                 [21, 22, 23, 24, 25, 26],
>                 [31, 32, 33, 34, 35, 36],
>                 [41, 42, 43, 44, 45, 46],
>                 [51, 52, 53, 54, 55, 56],
>                 ])
>     for i in range (3):
>         for j in range (4):
>             x2 [i][j] = x [i+1][j+0]
> 
> so, x2 is now what I want:
>     [[11,12,13,14,]
>      [21,22,23,24,]
>      [31,32,33,34,]]
> 
> I will be very happy to make the "slicing" without the loop...
> I'm pretty new to matlibplot, so sorry if this is a too simple question.
> 
> Thanks in advance for your help.
> 
>         Olivier
> -- 
>    . __    . ___  __.  | Olivier Bornet         Oli...@id...
>   / /  `  / /  / /  /  | IDIAP             http://www.idiap.ch/~bornet/
>  / /   / / /--/ /--'   | CP 592        http://www.idiap.ch/~bornet/pgp/
> / /__.' / /  / /       | CH-1920 Martigny           PGP-key: 0xC53D9218


`-