matplotlib-users Mailing List for matplotlib

Dear group

I am rather new to matplotlib. I have a reasonably large file of complex
valued data. Today I tried out specgram and psd, and they both work
nicely. However I would like the negative frequencies to be on the
negative side of the spectrum. I tried to look around, but could not
find any parameter to set this. My question is, if this is easily
possible or do I need to do something like

a,b,c,d=pylab.specgram(y,1024)
pylab.show()

x1,y1=a.shape
q=zeros([x1,y1],'d')
q[:,x1/2:x1-1]=a[:,0:x1/2-1]
q[:,0:x1/2-1]=a[:,x1-1:x1/2:-1]

pylab.imshow(q)

Apart from that I don't get the slicing to work properly yet, it would
be nicer if there was just a parameter to set for specgram



/Johan

Hi!

I'm trying to create a customized cursor crosshair in my plot. It is 
inserted into a gtk.Window. I'm using GTKAGG backend.

In order do this, I need to create a class CursorCrosshair for a code 
like this:

** CODE **

   win = gtk.Window()
   win.connect("destroy", lambda x: gtk.main_quit())
   win.set_default_size(400,300)

   figure = Figure()

   t = arange(0.0, 1.0, 0.01)
   s = sin(2*2*pi*t)

   sub1 = figure.add_subplot(111)
   sub1.plot(t, s, 'o')

   box = gtk.VBox()

   canvas = FigureCanvasGTKAgg(figure)
   cursor = CursorCrosshair()
   connect('motion_notify_event', cursor.draw_cursor)

   toolbar = NavigationToolbar2GTK( canvas, win )
   toolbar.update()

   box.pack_start(canvas, True, True)
   box.pack_end(toolbar, False, False)

   win.add(box)

   win.show_all()
   gtk.main()

*********


Into the draw_cursor method of CursorCrosshair class I'll paint the 
horizontal and vertical lines.

Well, the canvas in my example (FigureCanvasGTKAgg) inherits from 
gtk.DrawingArea, that is the GTK "real" canvas I need to paint into 
CursorCrosshair.draw_cursor(...) method.  gtk.DrawingArea has a method 
named draw_line(gc, x1, y1, x2, y2) that needs the coordinates and a 
graphics context as input parameters.

OK: how can I get the graphics context? What is it? And can I really use 
FigureCanvasGTKAgg like a gtk.DrawingArea object? If I cannot, how can I 
draw a line directly on the gtk inner canvas class (gtk.DrawingArea).

Thanks in advance.

Luigi

I just committed the changes to cvs and added a convenience function
for py2exe called get_py2exe_datafiles.  The script I pasted before
now looks like this:

Begin setup.py script
---------------------------------------------------------------------------=
-----------
# For py2exe only
"""
 Run with the following command (use py2exe 0.6.2 or higher)

   python.exe -OO setup.py py2exe -b 3 -c -p numarray,pytz -e numpy
"""

import os
from distutils.core import setup
import py2exe
import glob

import matplotlib

setup( version =3D '0.9.1',
       windows =3D ['nlogui.py'],
       data_files =3D [('', ['nlo.gif', '../vtkrotate/NMA.pdb']),
                     matplotlib.get_py2exe_datafiles()],
       options=3D{"py2exe":{"optimize":2}},
)

On 2/7/06, John Hunter <jdh...@ac...> wrote:
> >>>>> "Daniel" =3D=3D Daniel McQuillen <dan...@ya...> writes:
>
>     Daniel> Please be kind....poor starving newbie.  I've seen this
>     Daniel> question posted around but can't seem to find an answer:
>     Daniel> does anybody have experience creating an .exe for a
>     Daniel> matplotlib program using py2exe?
>
>     Daniel> My testMPL.py application is all set to go, but when I run
>     Daniel> py2exe with the suggested setup.py file, I get
>     Daniel> errors. (I'm using ActiveState Python 2.4.2 Build 10)
>
> Daniel, just for our information: are you using the py2exe examples
> from the matplotlib FAQ page?  I think these are probably a bit out of
> date as of the 0.86 release because of the way we recently reorganized
> the package data (fonts, thumbnails etc).
>
> Charlie, have you tested any of the new egg / package organization
> stuff with py2exe?

I just tried with an old project that I used py2exe with and it does
look like we still need the py2exe specific check in get_data_path
since py2exe zips the pure python code into a library.zip.  I updated
my setup.py file for that old project and I am pasting it below.  In
my specific case I was using numarray (numpy didn't exist), so now I
have to exclude numpy or errors occur for some unknown reason.  I just
added the old py2exe check to get_data_path and everything worked
fine.  I will add this to cvs.

Until the next release you can just uncomment the following lines in
matplotlib/__init__.py#_get_data_path():
    if sys.platform=3D=3D'win32' and sys.frozen:
        path =3D os.path.join(os.path.split(sys.path[0])[0], 'matplotlibdat=
a')
        if os.path.isdir(path):  return path
        else:
            # Try again assuming sys.path[0] is a dir not a exe
            path =3D os.path.join(sys.path[0], 'matplotlibdata')
            if os.path.isdir(path): return path

Begin setup.py script
---------------------------------------------------------------------------=
-----------
# For py2exe only
"""
 Run with the following command (use py2exe 0.6.2 or higher)

   python.exe -OO setup.py py2exe -b 3 -c -p numarray,pytz -e numpy
"""

import os
from distutils.core import setup
import py2exe
import glob

import matplotlib
mplfiles =3D glob.glob(os.sep.join([matplotlib.get_data_path(), '*']))
# Need to explicitly remove cocoa_agg nib folder or py2exe complains
mplfiles.remove(os.sep.join([matplotlib.get_data_path(), 'Matplotlib.nib'])=
)

setup( version =3D '0.9.1',
       windows =3D ['nlogui.py'],
       data_files =3D [('', ['nlo.gif', '../vtkrotate/NMA.pdb']),
                     ('matplotlibdata', mplfiles)],
       options=3D{"py2exe":{"optimize":2}},
)

Hi folks,

I have a question I couldn't solve so far. I'm trying to generate say 500=
 images
(png) without displaying them using imshow. I wrote a simple test functio=
n
(see below) which I call from a loop in which I create the data for the i=
mage in
a sequence. I systematically bump into a runtime error after saving about=
 188
files. After reading the (excellent) tutorial from Perry Greenfield and
Robert Jedrzejewski on interactive data analysis, I understand the need t=
o clean
up by freeing memory but I must be missing something else.

Cheers,

Aur=E9lien



def generateImshow(yarray,
                   xscansize,yscansize,
                   xscanstep,yscanstep,
                   outputfilename,
                   initimagesize =3D 10):
    '''Use this function to generate matplotlib images without displaying
them'''


    #reshape according to scan
    yarray =3D na.reshape(yarray,(yscansize,-1))

    #build image via matplotlib
    ximagesize =3D xscansize*xscanstep
    yimagesize =3D yscansize*yscanstep
    xyimageratio =3D float(ximagesize)/yimagesize
    print xyimageratio
    if xyimageratio > 1: ximagesize,yimagesize =3D
initimagesize,initimagesize*xyimageratio
    else: ximagesize,yimagesize =3D initimagesize*xyimageratio,initimages=
ize

    fig1 =3D pylab.figure(figsize=3D(ximagesize,yimagesize),dpi=3D100)
    #pylab.title('blahblah')
    im1 =3D pylab.imshow(yarray,
                 origin=3D'lower',
                 #aspect=3D'preserve',
                 #interpolation=3D'nearest',   #i.e. pixel
                 interpolation=3D'bicubic',   #i.e. smooth
                 #cmap=3Dmpl.cm.jet,          #not correct
                 #vmin=3Dminvalue,
                 #vmax=3Dmaxvalue
                 )
    pylab.colorbar()
    pylab.bone()
    pylab.axis('off')

    #save figure
    #outputfilename =3D filename[:-4]+'.png'
    pylab.savefig(outputfilename)
    #pylab.cla()
    del im1
    pylab.close('all')

>>>>> "andrea" == andrea gavana@tin it <and...@ti...> writes:

    andrea> Hello John & NG, thank you very much for your answer. It
    andrea> seems to me that using set_frame_on() just hides/shows the
    andrea> whole axes content.  Probably I didn't make myself clear
    andrea> (sorry for my bad english). I attach a jpeg example of
    andrea> what I mean for bon on/off. The figure at the left if with
    andrea> box "off", while at the right the box is "on".

Ahh, now I understand what you want.  Yes, this is a persistent
request and one we have not yet supported.  It is on the goals page,
however.  

JDH

>>>>> "Andrea" == Andrea Gavana <and...@ti...> writes:

    Andrea> Hello NG, I am having a couple of problems with Matplotlib
    Andrea> embedded in wxPython; both are related to the Pylab
    Andrea> twinx() function and its corresponding Matplotlib API
    Andrea> call.

Hi Andrea,

this is indeed a bug.  When we wrote support for the axis offsets, we
forgot about the use case where you might be using right ticks.

Darren, do you think this would be reasonably easy to support?

JDH

>>>>> "Daniel" == Daniel McQuillen <dan...@ya...> writes:

    Daniel> Please be kind....poor starving newbie.  I've seen this
    Daniel> question posted around but can't seem to find an answer:
    Daniel> does anybody have experience creating an .exe for a
    Daniel> matplotlib program using py2exe?

    Daniel> My testMPL.py application is all set to go, but when I run
    Daniel> py2exe with the suggested setup.py file, I get
    Daniel> errors. (I'm using ActiveState Python 2.4.2 Build 10)

Daniel, just for our information: are you using the py2exe examples
from the matplotlib FAQ page?  I think these are probably a bit out of
date as of the 0.86 release because of the way we recently reorganized
the package data (fonts, thumbnails etc).

Charlie, have you tested any of the new egg / package organization
stuff with py2exe?

JDH