matplotlib-users Mailing List for matplotlib (Page 8)

Hi,

just downloaded the last version of matplotlib 
(matplotlib-0.87.2.tar.gz) and the bug in setupext.py is still there... 
I AGAIN have to :
# add one line in setupext.py:

        o.tcl_inc="/usr/include/"

# before
        if not os.path.exists(o.tcl_inc):
            print 'cannot find tcl/tk headers. giving up.'
            sys.exit()
    return o


so that it works.

Is there a way that this bug could be corrected once and for all (or is 
it that I am doing something wrong?)

thanks
Eric

>>>>> "Darren" == Darren Dale <dd...@co...> writes:

    Darren> I have just started to think about this, and would
    Darren> appreciate any thoughts on how to approach the problem of
    Darren> interacting with data in the plot window. It seems like it
    Darren> might involve some pretty significant time and effort.

Have you seen examples/poly_editor.py

It shows you how to interactive add, delete and move vertices of a
polygon and can be used across backends (try launching the script with
-dTkAgg, -dGTKAgg etc),

The basic approach is to define a PolygonInteractor which is
initialized with a matplotlib.patches.Polygon and does all the event
connection and interaction for you.  The idea is to define an
Interactor for the import primitive objects (TextInteractor,
Line2DInteractor, etc) and then put these in an interactor module for
easy reuse.

There are other examples, not yet perfected (some edge and pixel
artifacts), which allow you to interactively create rectangles, move
them around and resize (grab on the corner and drag) using the
animation blit techniques.  I could dig them up if you are interested.

Another thing that is probably of general interest that I've been
talking to Charlie about off list is to develop a lasso tool for
selecting multiple objects.  

JDH

On 4/12/06, Darren Dale <dd...@co...> wrote:
> > I am interested in creating graphs where I can interact with the data
> >  plotted in the graph.

http://www.scipy.org/Wiki/Cookbook/Matplotlib/Interactive_Plotting

I submitted the above cookbook "recipe" a while back and I think it
may be an example of what you are looking for.   Although, it may be
kinda slow for large datasets.

I'm sure it could be made more elegant, or faster...  I'd be excited
to hear suggestions.

--chris

> I am interested in creating graphs where I can interact with the data
>  plotted in the graph.
>
>  For example, suppose I plot the heat of formation of ~ 80 different alloy
>  structures (which each have a name) against the composition of each
>  structure. What I would like is to hold the mouse over a data point and
> see a balloon pop up that displays the crystal structure name, or to left
> click on a data point to launch a structure viewer for the structure at
> that point, or right click to get a menu of different options, or to press
> "f" to see what filename the data for that point is saved in, etc...
>
>  Another example would be a graph of many lines, where each line
> corresponds to a structure. Then I would want to click on the lines to do
> various things.
>
>  Sometimes the data is very dense, so it would also be nice if you could
> move the focus to different points by tabbing or using the arrow buttons.
>
>  It would be interesting to be able to select portions of data points to do
>  some kind of analysis on, or to modify their appearance someway, maybe
>  change symbol the size according to atomic volume, or color according to
>  some property, etc.
> 
>  Could these kinds of things be done in/with matplotlib? Or more
> importantly, does a framework in matplotlib exist that this kind of thing
> could be developed? I am interested in talking to anyone who has thoughts
> about this.

I use a really nice application called PyMca at work for analyzing x-ray 
fluorescence spectra. The gui is built with PyQt, and the author is 
considering replacing PyQwt with matplotlib. PyMca also needs to be able to 
interact with the plot, specifically to draw markers or lines and move them 
around with the mouse.

I have just started to think about this, and would appreciate any thoughts on 
how to approach the problem of interacting with data in the plot window. It 
seems like it might involve some pretty significant time and effort.

Darren

Hi All,=0A=
=0A=
I'm having a problem using Affines to generate a rotated XY plot. The=0A=
code below works with the exception of the ax.text(...) statement which=0A=
doesn't put the text where I think it should go.=0A=
=0A=
Anyone know what I'm doing wrong?=0A=
=0A=
Thanks,=0A=
Alan.=0A=
=0A=
from pylab import figure, show=0A=
from matplotlib.numerix import arange=0A=
from matplotlib.transforms import Value, Affine=0A=
=0A=
line_props =3D dict(linewidth=3D0.5, color=3D'black', linestyle=3D'-')=0A=
fig =3D figure()=0A=
ax =3D fig.add_axes([0.1, 0.1, 0.8, 0.8])=0A=
=0A=
# Create Affine with 45 degree rotation=0A=
temp_orig =3D Value(-20)=0A=
theta_orig =3D Value(-20)=0A=
temp_width =3D Value(120)=0A=
=0A=
scale =3D (ax.bbox.ur().x() - ax.bbox.ll().x())/temp_width=0A=
tx =3D ax.bbox.ll().x() - (temp_orig*scale + theta_orig*scale)=0A=
ty =3D ax.bbox.ll().y() - (temp_orig*scale - theta_orig*scale)=0A=
affine =3D Affine(scale, Value(0)-scale, scale, scale, tx, ty)=0A=
=0A=
# Plot (rotated) gridlines=0A=
temp_min =3D -70=0A=
temp_max =3D 40=0A=
theta_min =3D -20=0A=
theta_max =3D 90=0A=
=0A=
temp =3D arange(temp_min, temp_max, 10)=0A=
theta =3D arange(theta_min, theta_max, 10)=0A=
=0A=
for t in temp:=0A=
    ax.plot((t, t), (theta_min, theta_max), transform=3Daffine,=0A=
**line_props)=0A=
=0A=
for t in theta:=0A=
    ax.plot((temp_min, temp_max), (t, t), transform=3Daffine,=0A=
**line_props)=0A=
=0A=
# Plot a small square around (0,20) - this works OK=0A=
ax.plot((-1, -1, 1, 1, -1), (19, 21, 21, 19, 19), transform=3Daffine)=0A=
=0A=
# Plot text at (0, 20)- should be in the same place as the small square,=0A=
but isn't=0A=
ax.text(0, 20, 'Test', transform=3Daffine)=0A=
=0A=
show()=0A=
=0A=
--=20=0A=
=0A=
Visit our website at www.roke.co.uk=0A=
=0A=
Roke Manor Research Ltd, Roke Manor, Romsey, Hampshire SO51 0ZN, UK.=0A=
=0A=
The information contained in this e-mail and any attachments is proprietary=
 to=0A=
Roke Manor Research Ltd and must not be passed to any third party without=
=0A=
permission. This communication is for information only and shall not create=
 or=0A=
change any contractual relationship.=0A=
=0A=

Hello NG,

    I haven't stated my question very clearly, probably. I 
am trying to use imshow() with a dinamically changing set of matrices.
What my application does, is just to respond to a user selection from 
awxPython radiobutton and, based on the user's choice, display a 
different matrix on a matplotlib figure (I am using the OO interface). 
I also have an horizontal colorbar to let the user orient himself over 
the matrix values. Well, these matrices have different magnitues in 
value, and I was expecting that the colorbar would be updated with the 
new image data. 
Isn't the colorbar supposed to update when the input 
image/pcolor/whatever changes?

I attach a modified example of 
dynamic_image_wxagg2.py that demonstrates the problem: the values in 
the image increase, but the colorbar stays quiet at its bounds 
(0.0,2.0) and ends up filled only with blue...

Does anyone have a 
suggestion?

Thank you.

Andrea.

* Robert Kern
  <rob...@gm...>
wrote:
> Fabian Braennstroem wrote:
>> Hi Robert,
>> 
>> * Robert Hetland <he...@ta...>
>>   wrote:
>> 
>>>On Apr 11, 2006, at 11:27 AM, Robert Kern wrote:
>>>
>>>>>In my ubuntu 0.3.2 and archlinux 0.4.8 version there is no
>>>>>such delauny directory. There exists a sandbox directory,
>>>>>but this looks like:
>>>>>
>>>>>__init__.py  __init__.pyc  __init__.pyo  build  setup.py
>>>>>setup.pyc  setup.pyo
>>>>>
>>>>>Do I miss any package or should I download it by hand from
>>>>>http://projects.scipy.org/scipy/scipy/browser/trunk/Lib/sandbox
>>>
>>>This is in the scipy source, not in the scipy directory installed in  
>>>python/lib.  This should be downloaded by default if you download  
>>>from svn.  BTW, kudos to Robert Kern for making this excellent  
>>>griding package.
>> 
>> I could install the package using the svn version, thanks!
>> But now, I am not sure, if it works. It seems for me that it
>> is a different problem compared to Nils. I get:
>> 
>> `--> python gridding_matpl.py
>> /usr/lib/python2.4/site-packages/matplotlib/__init__.py:719: UserWarning: Bad val "numpy"                                                                                            on line #33
>>         "numerix      : numpy     #Numeric   # Numeric or numarray"
>>         in file "/etc/matplotlibrc"
>>         Numerix must be Numeric or numarray
>
> You have an old version of matplotlib. I would suggest upgrading if you want to
> use numpy with matplotlib.

Thanks, I just thought that my dapper/ubuntu is a little bit
fresher ... I will try it in the next days and let you know.

Greetings!
 Fabian

Fabian Braennstroem wrote:
> Hi Robert,
> 
> * Robert Hetland <he...@ta...>
>   wrote:
> 
>>On Apr 11, 2006, at 11:27 AM, Robert Kern wrote:
>>
>>>>In my ubuntu 0.3.2 and archlinux 0.4.8 version there is no
>>>>such delauny directory. There exists a sandbox directory,
>>>>but this looks like:
>>>>
>>>>__init__.py  __init__.pyc  __init__.pyo  build  setup.py
>>>>setup.pyc  setup.pyo
>>>>
>>>>Do I miss any package or should I download it by hand from
>>>>http://projects.scipy.org/scipy/scipy/browser/trunk/Lib/sandbox
>>
>>This is in the scipy source, not in the scipy directory installed in  
>>python/lib.  This should be downloaded by default if you download  
>>from svn.  BTW, kudos to Robert Kern for making this excellent  
>>griding package.
> 
> I could install the package using the svn version, thanks!
> But now, I am not sure, if it works. It seems for me that it
> is a different problem compared to Nils. I get:
> 
> `--> python gridding_matpl.py
> /usr/lib/python2.4/site-packages/matplotlib/__init__.py:719: UserWarning: Bad val "numpy"                                                                                            on line #33
>         "numerix      : numpy     #Numeric   # Numeric or numarray"
>         in file "/etc/matplotlibrc"
>         Numerix must be Numeric or numarray

You have an old version of matplotlib. I would suggest upgrading if you want to
use numpy with matplotlib.

-- 
Robert Kern
rob...@gm...

"I have come to believe that the whole world is an enigma, a harmless enigma
 that is made terrible by our own mad attempt to interpret it as though it had
 an underlying truth."
  -- Umberto Eco

Gary writes:
 > Ryan Krauss wrote:
 > 
 > >I will test out a 600dpi png in a powerpoint slide and see how I feel about it.
 > >
 > Would OpenOffice.Org work better in this regard than PP ??  E.g., 
 > display eps.
 > (never tried it myself)

My experience has been that any character that belongs on a
presentation slide (i.e. that will be visible from a projector screen)
looks ok when converted to a  300 dpi bitmap, with the image reduced
and cropped in powerpoint.  ImageMagick does a great job:

convert -density 300 -trim file.eps file.gif

Regards, Phil

Hi Robert,

* Robert Hetland <he...@ta...>
  wrote:
>
> On Apr 11, 2006, at 11:27 AM, Robert Kern wrote:
>>> In my ubuntu 0.3.2 and archlinux 0.4.8 version there is no
>>> such delauny directory. There exists a sandbox directory,
>>> but this looks like:
>>>
>>> __init__.py  __init__.pyc  __init__.pyo  build  setup.py
>>> setup.pyc  setup.pyo
>>>
>>> Do I miss any package or should I download it by hand from
>>> http://projects.scipy.org/scipy/scipy/browser/trunk/Lib/sandbox
>
> This is in the scipy source, not in the scipy directory installed in  
> python/lib.  This should be downloaded by default if you download  
> from svn.  BTW, kudos to Robert Kern for making this excellent  
> griding package.

I could install the package using the svn version, thanks!
But now, I am not sure, if it works. It seems for me that it
is a different problem compared to Nils. I get:

`--> python gridding_matpl.py
/usr/lib/python2.4/site-packages/matplotlib/__init__.py:719: UserWarning: Bad val "numpy"                                                                                            on line #33
        "numerix      : numpy     #Numeric   # Numeric or numarray"
        in file "/etc/matplotlibrc"
        Numerix must be Numeric or numarray
  warnings.warn('Bad val "%s" on line #%d\n\t"%s"\n\tin file "%s"\n\t%s' % (val, cnt, line                                                                                           , fname, msg))
Traceback (most recent call last):
  File "gridding_matpl.py", line 43, in ?
    plot_data(xi,yi,zi)
  File "gridding_matpl.py", line 16, in plot_data
    pcolor(xi,yi,zim,shading='interp',cmap=cm.gray)
  File "/usr/lib/python2.4/site-packages/matplotlib/pylab.py", line 1941, in pcolor
    ret =  gca().pcolor(*args, **kwargs)
  File "/usr/lib/python2.4/site-packages/matplotlib/axes.py", line 2142, in pcolor
    C = ma.asarray(C)
  File "/usr/lib/python2.4/site-packages/Numeric/MA/MA.py", line 2090, in asarray
    return array(data, typecode=typecode, copy=0)
  File "/usr/lib/python2.4/site-packages/Numeric/MA/MA.py", line 591, in __init__
    c = Numeric.array(data, tc, savespace=ss)
  File "/usr/lib/python2.4/site-packages/numpy/core/ma.py", line 613, in __array__
    raise MAError, \
numpy.core.ma.MAError: Cannot automatically convert masked array to numeric because data
                          is masked in one or more locations.

As you see, I am not 'allowed' to change 'numerix' to
'numpy' and printing the values from 'zi' with 'print zi' gives me:

[[        nan         nan         nan ...,         nan         nan
          nan]
 [        nan         nan         nan ...,         nan         nan
          nan]
 [        nan         nan         nan ...,         nan         nan
          nan]
 ..., 
 [        nan         nan         nan ...,         nan         nan
          nan]
 [        nan         nan         nan ...,         nan         nan
          nan]
 [        nan         nan         nan ...,         nan         nan
          nan]]


Sorry, but do you have some more suggestions?

Greetings!
 Fabian

Ryan Krauss wrote:

>I will test out a 600dpi png in a powerpoint slide and see how I feel about it.
>
Would OpenOffice.Org work better in this regard than PP ??  E.g., 
display eps.
(never tried it myself)

Hi,

From other postings I see how to use a legend with errorbar.  For example:
 http://sourceforge.net/mailarchive/message.php?msg_id=14124668

But I see that there is still a problem with the *color* of the legend
symbols.  Specifically, if I plot using errorbar and then overplot a line,
the color of the line in the legend is set to the color of the errorbar
plot.

For example:

from pylab import *
data = [1,2,3]
error = [.5, .5, .5]
# The errorbar plot is blue.
line1 = errorbar(data, data, error, None, 'bo')
# The dashed line plot is red.
line2 = plot(data, data, 'r--')
legend((line1,line2),('line1','line2'))
show()

Notice that in the legend both the points and the dashed lines are blue,
when the dashed line should be red!

Any ideas on how to get around this, or better yet, patch it?

Thanks,
James

On Apr 11, 2006, at 11:27 AM, Robert Kern wrote:
>> In my ubuntu 0.3.2 and archlinux 0.4.8 version there is no
>> such delauny directory. There exists a sandbox directory,
>> but this looks like:
>>
>> __init__.py  __init__.pyc  __init__.pyo  build  setup.py
>> setup.pyc  setup.pyo
>>
>> Do I miss any package or should I download it by hand from
>> http://projects.scipy.org/scipy/scipy/browser/trunk/Lib/sandbox

This is in the scipy source, not in the scipy directory installed in  
python/lib.  This should be downloaded by default if you download  
from svn.  BTW, kudos to Robert Kern for making this excellent  
griding package.

-Rob

-----
Rob Hetland, Assistant Professor
Dept of Oceanography, Texas A&M University
p: 979-458-0096, f: 979-845-6331
e: he...@ta..., w: http://pong.tamu.edu

Fabian Braennstroem wrote:
> Hi,
> 
> * Robert Hetland <he...@ta...>
>   wrote:
> 
>>It's in the sandbox:
>>
>>scipy/Lib/sandbox/delaunay
>>
>>Go to this directory, and install with the usual python setup.py  
>>build; python setup.py install
> 
> In my ubuntu 0.3.2 and archlinux 0.4.8 version there is no
> such delauny directory. There exists a sandbox directory,
> but this looks like:
> 
> __init__.py  __init__.pyc  __init__.pyo  build  setup.py
> setup.pyc  setup.pyo
> 
> Do I miss any package or should I download it by hand from 
> http://projects.scipy.org/scipy/scipy/browser/trunk/Lib/sandbox

All of the sandbox subpackages are turned off by default because they are in
development and may not build. You can turn them on by uncommenting or adding
the appropriate line in scipy/Lib/sandbox/setup.py and rebuilding scipy.

-- 
Robert Kern
rob...@gm...

"I have come to believe that the whole world is an enigma, a harmless enigma
 that is made terrible by our own mad attempt to interpret it as though it had
 an underlying truth."
  -- Umberto Eco

Robert Hetland wrote:
> 
> It's in the sandbox:
> 
> scipy/Lib/sandbox/delaunay
> 
> Go to this directory, and install with the usual python setup.py  build;
> python setup.py install
> 
> I'll change the wiki so this is more clear.

The preferred way to build packages in the sandbox is to uncomment (or add) the
appropriate line in scipy/Lib/sandbox/setup.py and rebuild scipy. They should
then be referred to as scipy.sandbox.delaunay, for example.

-- 
Robert Kern
rob...@gm...

"I have come to believe that the whole world is an enigma, a harmless enigma
 that is made terrible by our own mad attempt to interpret it as though it had
 an underlying truth."
  -- Umberto Eco

Hi,

* Robert Hetland <he...@ta...>
  wrote:
>
> It's in the sandbox:
>
> scipy/Lib/sandbox/delaunay
>
> Go to this directory, and install with the usual python setup.py  
> build; python setup.py install

In my ubuntu 0.3.2 and archlinux 0.4.8 version there is no
such delauny directory. There exists a sandbox directory,
but this looks like:

__init__.py  __init__.pyc  __init__.pyo  build  setup.py
setup.pyc  setup.pyo

Do I miss any package or should I download it by hand from 
http://projects.scipy.org/scipy/scipy/browser/trunk/Lib/sandbox

Greetings!
Fabian

Powerpoint supports PNG & TIFF

One option would thus be to generate an EPS with the PS backend (using
usetex)

Then convert it to TIFF of PNG using Ghostscript.

Here is an example from a python script I use to generate png's with
alpha transparance for web use with M4 Circuit Macros...=20

...
pngwrite_cmd=3D"gswin32c -dQUIET -dBATCH -dNOPAUSE -dSAFER -dEPSCrop
-sDEVICE=3Dpngalpha -r600 -sOutputFile=3D"+filename+".png =
"+filename+".eps"
error=3Dos.system(pngwrite_cmd)
...

Then in PowderPoint use "Insert" -> "Picture" -> Graphics from File"=20

And "drag" to the desired size...

PJR


> -----Original Message-----
> From: mat...@li...
> [mailto:mat...@li...] On=20
> Behalf Of Ryan Krauss
> Sent: 11 April 2006 16:39
> To: matplotlib-users
> Subject: Re: [Matplotlib-users] matplotlib plots into powerpoint
>=20
>=20
> I will test out a 600dpi png in a powerpoint slide and see
> how I feel about it.
>=20
> I need to use usetex.  Does that work with the svg backend?
>=20
> As a last resort, I will dump the data to a text file from
> Python and use Matlab to generate wmf's (the horror!).  I am=20
> o.k. with it because this is the last time I will ever have=20
> to do one of these (provided I graduate soon).
>=20

Ryan Krauss wrote:
> I will test out a 600dpi png in a powerpoint slide and see how I feel about it.

You can always resize/rescale it...
I'm puzzled.

m.
-- 
Massimo Sandal
University of Bologna
Department of Biochemistry "G.Moruzzi"

snail mail:
Via Irnerio 48, 40126 Bologna, Italy

email:
mas...@un...

tel: +39-051-2094388
fax: +39-051-2094387

On Tuesday 11 April 2006 10:39 am, Ryan Krauss wrote:
> I will test out a 600dpi png in a powerpoint slide and see how I feel about
> it.
>
> I need to use usetex.  Does that work with the svg backend?

No.

On Tue, 11 Apr 2006, Ryan Krauss apparently wrote:=20
> I have to put some matplotlib plots into a poster in=20
> Powerpoint that will be printed out for a research review=20
> (my advisor requires the use of Powerpoint).  I would=20
> prefer not to go the high dpi jpg or png route, since=20
> these are going to be printed.  And Powerpoint won't=20
> actually show you the graph on the screen if you insert=20
> eps or pdf.  I think my best bet is wmf.  Should I be=20
> considering some other format?=20

http://www.cis.upenn.edu/~mwh/tips.html
http://duramecho.com/ComputerInformation/HowToDoPowerpointPosters.html

hth,
Alan Isaac

It's in the sandbox:

scipy/Lib/sandbox/delaunay

Go to this directory, and install with the usual python setup.py  
build; python setup.py install

I'll change the wiki so this is more clear.

-Rob.




On Apr 11, 2006, at 8:34 AM, Fabian Braennstroem wrote:

> Hi,
>
> * Andrew Straw
>   <str...@as...> wrote:
>> Nils Wagner wrote:
>>
>>> Hi all,
>>>
>>> Given a set of points (x_i,y_i,z_i)
>>> is it possible to plot y_i versus x_i with a varying color
>>> corresponding to the magnitude of z_i ?
>>>
>>> Nils
>>>
>> http://scipy.org/Wiki/Cookbook/Matplotlib/ 
>> Gridding_irregularly_spaced_data
>
> Maybe, I am kind of stupid, but where do I find the delaunay
> module used in the example on that side?
>
> Greetings!
>  Fabian
>
>
>
> -------------------------------------------------------
> This SF.Net email is sponsored by xPML, a groundbreaking scripting  
> language
> that extends applications into web and mobile media. Attend the  
> live webcast
> and join the prime developer group breaking into this new coding  
> territory!
> http://sel.as-us.falkag.net/sel? 
> cmd=lnk&kid=110944&bid=241720&dat=121642
> _______________________________________________
> Matplotlib-users mailing list
> Mat...@li...
> https://lists.sourceforge.net/lists/listinfo/matplotlib-users

-----
Rob Hetland, Assistant Professor
Dept of Oceanography, Texas A&M University
p: 979-458-0096, f: 979-845-6331
e: he...@ta..., w: http://pong.tamu.edu

I will test out a 600dpi png in a powerpoint slide and see how I feel about=
 it.

I need to use usetex.  Does that work with the svg backend?

As a last resort, I will dump the data to a text file from Python and
use Matlab to generate wmf's (the horror!).  I am o.k. with it because
this is the last time I will ever have to do one of these (provided I
graduate soon).

On 4/11/06, Gary Ruben <gr...@bi...> wrote:
> Hi Ryan,
> You could try the emf backend directly from matplotlib. I just had my
> first go at using it and got a traceback. Then I put
> import matplotlib
> matplotlib.use('emf')
> before the pylab import as suggested here:
> http://matplotlib.sourceforge.net/backends.html
> and it worked, although mathtext doesn't seem to work with it. It also
> produced a very jagged line in my test plot, where other backends
> produce nice smooth ones, so I'm not convinced I'd use it, but it might
> be OK for what you want.
> Gary R.
>
> Ryan Krauss wrote:
> > I have to put some matplotlib plots into a poster in Powerpoint that
> > will be printed out for a research review (my advisor requires the use
> > of Powerpoint).  I would prefer not to go the high dpi jpg or png
> > route, since these are going to be printed.  And Powerpoint won't
> > actually show you the graph on the screen if you insert eps or pdf.  I
> > think my best bet is wmf.  Should I be considering some other format?
> > If wmf is my best bet, how do I get my plots from matplotlib into wmf?
> >  I thought I remembered a wmf output option from ghostscript on
> > windows, but I just installed gs on my wife's windows  computer and
> > don't see it.
> >
> > Thanks,
> >
> > Ryan
>
>

Hi Ryan,
You could try the emf backend directly from matplotlib. I just had my 
first go at using it and got a traceback. Then I put
import matplotlib
matplotlib.use('emf')
before the pylab import as suggested here:
http://matplotlib.sourceforge.net/backends.html
and it worked, although mathtext doesn't seem to work with it. It also 
produced a very jagged line in my test plot, where other backends 
produce nice smooth ones, so I'm not convinced I'd use it, but it might 
be OK for what you want.
Gary R.

Ryan Krauss wrote:
> I have to put some matplotlib plots into a poster in Powerpoint that
> will be printed out for a research review (my advisor requires the use
> of Powerpoint).  I would prefer not to go the high dpi jpg or png
> route, since these are going to be printed.  And Powerpoint won't
> actually show you the graph on the screen if you insert eps or pdf.  I
> think my best bet is wmf.  Should I be considering some other format? 
> If wmf is my best bet, how do I get my plots from matplotlib into wmf?
>  I thought I remembered a wmf output option from ghostscript on
> windows, but I just installed gs on my wife's windows  computer and
> don't see it.
> 
> Thanks,
> 
> Ryan

>>>>> "Ryan" == Ryan Krauss <rya...@gm...> writes:

    Ryan> I have to put some matplotlib plots into a poster in
    Ryan> Powerpoint that will be printed out for a research review
    Ryan> (my advisor requires the use of Powerpoint).  I would prefer

Please purchase a copy of "The Cognitive Style of Powerpoint" by
Edward Tufte

  http://www.edwardtufte.com/tufte/books_pp

    Ryan> not to go the high dpi jpg or png route, since these are
    Ryan> going to be printed.  And Powerpoint won't actually show you

Are you worried about the printing being slow?  As long as the dpi is
high enough, the quality should suffice.  Does powerpoint handle svg?
The matplotlib svg backend is reasonably complete (no mathtext).

JDH

I have to put some matplotlib plots into a poster in Powerpoint that
will be printed out for a research review (my advisor requires the use
of Powerpoint).  I would prefer not to go the high dpi jpg or png
route, since these are going to be printed.  And Powerpoint won't
actually show you the graph on the screen if you insert eps or pdf.  I
think my best bet is wmf.  Should I be considering some other format?=20
If wmf is my best bet, how do I get my plots from matplotlib into wmf?
 I thought I remembered a wmf output option from ghostscript on
windows, but I just installed gs on my wife's windows  computer and
don't see it.

Thanks,

Ryan