matplotlib-users Mailing List for matplotlib (Page 2)

>>>>> "K" == K KISHIMOTO <ko...@us...> writes:

    K> Hi, I'm using matplotlib for my scientific analysis, and
    K> charmed by It's function.

Thanks!

    K> Because I often use another tools for data histograming, I
    K> think it's convenient if matplotlib could plot already
    K> histogramed data more efficiently.

This seems like the kind of thing that would best be done in your user
library.  Eg, if you make a module mymatplotlib.py, you can defined
your own hist.  In that file, just import matplotlib.matlab and call
matplotlib.matlab.hist within it.  In fact, matplotlib.matlab.hist
calls matplotlib.mlab.hist.

On an unrelated topic: 

  def hist(x, bins=10, noplot=0, normed=0, weights=[], errors=[])

Just wanted to point out that this is a potential gotcha in python.
For example, what do you expect the output of this code to be?

def func1(x=[]):
    if not len(x): x.append(1)
    print x
    
def func2(y=[]):
    if len(y)<2: y.append(2)
    print y

z = []
func1(z)
func2(z)

The standard way to pass empty lists as default function args is to
do:

def func1(x=None):
    if x is None: x = []
    #blah blah    

Lists and dicts are different in this capacity than strings or ints
because the can be changed (mutable).

Thanks for your suggestions,
John Hunter

Hi,

I'm using matplotlib for my scientific analysis, and charmed by It's 
function.

Because I often use another tools for data histograming,
I think it's convenient if matplotlib could plot already histogramed 
data more efficiently.

For exmple, changing from
hist(x, bins=10, noplot=0, normed=0)
to
hist(x, bins=10, noplot=0, normed=0, weights=[], errors=[])

Calling
hist([0.5, 1.5, 1.7, 2.5], [0.0, 1.0, 3.0])
equals to
hist([0.5, 1.6, 2.5], [0.0, 1.0, 3.0], weights=[1.0, 2.0, 0.0])
or
hist([0.5, 1.6, 2.5], [0.0, 1.0, 3.0], weights=[1.0, 2.0, 0.0], 
errors=[1.0, 1.414, 0.0])

(1) When "weights" argument is provided, "x" is assumed as a list of 
weighted mean (or center) values of each bin and "weights" is assumed 
as a list of (already histogramed) weights of each bin.
(2) When "errors" argument is omitted, it's default values are set to 
the squares of weights.
(3) When "errors" argument is provided, it's values are used as a list 
of one-sigma errors of each bin.

Thank you for reading.
Koji

>>>>> "Carlos" == Carlos  <ra...@ho...> writes:

    Carlos> Hi whats a good way to add a file open dialog box to a
    Carlos> matplotlib script like the one in the demo called
    Carlos> "simple_plot.py". I'm new to python and am really
    Carlos> impressed with using matplotlib. Is there an EZ way to add
    Carlos> a file select dialog?

Yes.

If you are using the GTK backend, there are two examples in the
examples dir of the source distribution (*.zip or *.tar.gz) to help
you out.

  1) embedding_in_gtk2.py shows you how to use matplotlib in a GTK example.

  2) mpl_with_glade.glade and mpl_with_glade.py show you how to use
     matplotlib with gtk and glade.  glade is a RAD (rapid application
     development) tool that allows you to design GUIs in a
     drag-and-drop environment.

To use either of these you'll still have to learn pygtk
http://www.daa.com.au/~james/pygtk/ and/or glade
http://glade.gnome.org.

If you want to use the wx backend, jeremy has provided an example of
how to use matplotlib in a wx application embedding_in_wx.py.

While none of these examples specifically address the issue of file
selection dialogs, if you know something about the backend GUI widget
library, it's easy to add File Selection to the examples provided.  

JDH

Hi=20
    whats a good way to add a file open dialog box to a matplotlib =
script like the one in the demo called "simple_plot.py". I'm new to =
python and am really impressed with using matplotlib. Is there an EZ way =
to add a file select dialog?

_____________________________

>>>>> "LUK" == LUK ShunTim <shu...@po...> writes:

    LUK> I had matplotlib installed in another location, not under the
    LUK> default python main tree and use the environmental variable
    LUK> AFMPATH to point to the afm fonts files. The backend_wx(.py)
    LUK> relies on distutils.sysconfig.PREFIX to find the GUI bitmap
    LUK> files. Adding the following 2 lines

A number of people have reported problems with matplotlib in
non-default install locations.  I'll try and get these cleared up
before the next release.

Thanks! 
John Hunter

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

    Flavio> some other example fail with the exact same traceback... I
    Flavio> believe the problem lies with the gtk backend, since the
    Flavio> same example runs fine with the WX backend.

Your error message suggests to me that you have an out-of-day CVS
copy, which is not surprising since sourceforge seems to have problem
staying current.

Grab another copy when you get the chance and let me know if you have
any more problems.

JDH

Hello,

I just tried out the wxpython backend from the CVS and it seems to work 
as advertised, except a small glitch.

I had matplotlib installed in another location, not under the default 
python main tree and use the environmental variable AFMPATH to point to 
the afm fonts files. The backend_wx(.py) relies on 
distutils.sysconfig.PREFIX to find the GUI bitmap files. Adding the 
following 2 lines

     if os.environ.has_key('AFMPATH'):
         basedir = os.environ['AFMPATH']

after the line

basedir = os.path.join(distutils.sysconfig.PREFIX, 'share', 'matplotlib')

in the function  _load_bitmap(filename) will enable it to locate the 
bitmap files which are installed together with the afm font files. (As 
AFMPATH points to more than font resources, a more appropriate name such 
as MATPLOTLIBRESOURCE may be warranted, though changing it will break 
old installations. I'm happy to let it stay as it is.)

Regards,
ST

Hi Flavio,

I expect that you've found the solution to your problem now, but just in
case, I have now published significantly improved version of the wx
embedding example in CVS.

Since Sourceforge can take some time to sort these things out, I have
included the code here.

I should also note that there has been quite a significant enhancement to
the CVS version of backend_wx committed to CVS. You may wish to use this
(although I don't think that the interface has changed anywhere.

Details of the changes have been posted to the matplotlib-devel list, or
you can see details at the top of the source file (matplotlib-devel
archiving appears to be very slow).

Regards

Jeremy

=3D3D=3D3D=3D3D=3D3D=3D3D=3D3D=3D3D=3D3D=3D3D=3D3D=3D3D=3D3D=3D3D=3D3D=3D=
3D=3D3D=3D3D=3D3D=3D3D

"""
Copyright (C) Jeremy O'Donoghue, 2003

License: This work is licensed under the PSF. A copy should be included
with this source code, and is also available at
http://www.python.org/psf/license.html

This is a sample showing how to embed a matplotlib figure in a wxPanel.

The example implements the full navigation toolbar, so you can automatica=
=3D
lly
inherit standard matplotlib features such as the ability to zoom, pan and
save figures in the supported formats.

There are a few small complexities worth noting in the example:

1) By default, a wxFrame can contain a toolbar (added with SetToolBar())
   but this is at the top of the frame. Matplotlib default is to put the
   controls at the bottom of the frame, so you have to manage the toolbar
   yourself. I have done this by putting the figure and toolbar into a
   sizer, but this means that you need to override GetToolBar for your
   wxFrame so that the figure manager can find the toolbar.

2) I have implemented a figure manager to look after the plots and axes.
   If you don't want a toolbar, it is simpler to add the figure directly
   and not worry. However, the figure manager looks after clipping of the
   figure contents, so you will need it if you want to navigate

3) There is a bug in the way in which my copy of wxPython calculates
   toolbar width on Win32, so there is a tricky line to ensure that the
   width of the toolbat is the same as the width of the figure.

4) Depending on the parameters you pass to the sizer, you can make the
   figure resizable or not.
"""

import matplotlib
matplotlib.use('WX')

from matplotlib.backends import Figure, Toolbar, FigureManager
from matplotlib.axes import Subplot
import Numeric as numpy

from wxPython.wx import *

class PlotFigure(wxFrame):
    def __init__(self):
        wxFrame.__init__(self, None, -1, "Test embedded wxFigure")
        self.fig =3D3D Figure(self, -1, (5,4), 75)
        self.toolbar =3D3D Toolbar(self.fig)
        self.toolbar.Realize()

                # On Windows, default frame size behaviour is incorrect
        # you don't need this under Linux
        tw, th =3D3D self.toolbar.GetSizeTuple()
        fw, fh =3D3D self.fig.GetSizeTuple()
        self.toolbar.SetSize(wxSize(fw, th))

        # Create a figure manager to manage things
        self.figmgr =3D3D FigureManager(self.fig, 1, self)

        # Now put all into a sizer
        sizer =3D3D wxBoxSizer(wxVERTICAL)
                # This way of adding to sizer prevents resizing
        #sizer.Add(self.fig, 0, wxLEFT|wxTOP)

                # This way of adding to sizer allows resizing
        sizer.Add(self.fig, 1, wxLEFT|wxTOP|wxGROW)

                # Best to allow the toolbar to resize!
        sizer.Add(self.toolbar, 0, wxGROW)
        self.SetSizer(sizer)
        self.Fit()

    def plot_data(self):
        # Use ths line if using a toolbar
        a =3D3D self.figmgr.add_subplot(111)

        # Or this one if there is no toolbar
        #a =3D3D Subplot(self.fig, 111)

        t =3D3D numpy.arange(0.0,3.0,0.01)
        s =3D3D numpy.sin(2*numpy.pi*t)
        c =3D3D numpy.cos(2*numpy.pi*t)
        a.plot(t,s)
        a.plot(t,c)
        self.toolbar.update()

    def GetToolBar(self):
        # You will need to override GetToolBar if you are using an
        # unmanaged toolbar in your frame
        return self.toolbar

if __name__ =3D3D=3D3D '__main__':
    app =3D3D wxPySimpleApp()
    frame =3D3D PlotFigure()
    frame.plot_data()
    frame.Show()
    app.MainLoop()