matplotlib-users Mailing List for matplotlib (Page 3)

Great. Thanks. I'll take a look at that file. Is "transFigure" the
one that I was calling "screen space"?
--bb
On 7/28/06, Eric Firing <ef...@ha...> wrote:
> Bill,
>
> You can see an example of how to do something similar by looking at the
> QuiverKey class in quiver.py. It is all a matter of using the
> transforms module.
>
> Eric
>
> Bill Baxter wrote:
> > I want to draw some labels with plot.text() and have them appear a
> > given number of pixels (or mm, or points) to above and to the right of
> > the data points they are describing. Is there some way to specify a
> > screen offset from a point in graph coordinates? Like a method of the
> > axes that converts from screen to graph coords? Using graph
> > coordinates means that when I zoom in really close to see some
> > details, the text labels are way out in never-never land.
> >
> > Thanks,
> > --bb
>

Hello,
 I am trying to install matplotlib-0.87.3 on a Fedora Core 4 box
The system has python-2.4 installed
All of the necessary addons- scipy, numarray, Numeric, gtk, etc have
been added.
I am including the output of running
$ python setup.py install
Any help is appreciated.
-Thanks,
 -sen1
Output of above script:
/usr/lib/python2.4/site-packages/numpy/lib/utils.py:92:
DeprecationWarning: get_
numpy_include is deprecated, use get_include
 DeprecationWarning)
running install
running build
running build_py
running build_ext
building 'matplotlib._ns_cntr' extension
C compiler: gcc -pthread -fno-strict-aliasing -DNDEBUG -O2 -g -pipe
-Wp,-D_FORTI
FY_SOURCE=3D2 -fexceptions -m32 -march=3Di386 -mtune=3Dpentium4
-fasynchronous-unwind-
tables -D_GNU_SOURCE -fPIC -fPIC
compile options:
'-I/usr/lib/python2.4/site-packages/numpy/core/include -I/usr/l
ocal/include -I/usr/include -I. -I/usr/include/python2.4 -c'
extra options: '-DSCIPY=3D1'
gcc: src/_ns_cntr.c
src/_ns_cntr.c: In function =E2=80=98Cntr_init=E2=80=99:
src/_ns_cntr.c:1582: error: =E2=80=98PyArray_SBYTE=E2=80=99 undeclared (f=
irst use in
this functi
on)
src/_ns_cntr.c:1582: error: (Each undeclared identifier is reported
only once
src/_ns_cntr.c:1582: error: for each function it appears in.)
src/_ns_cntr.c: In function =E2=80=98Cntr_init=E2=80=99:
src/_ns_cntr.c:1582: error: =E2=80=98PyArray_SBYTE=E2=80=99 undeclared (f=
irst use in
this functi
on)
src/_ns_cntr.c:1582: error: (Each undeclared identifier is reported
only once
src/_ns_cntr.c:1582: error: for each function it appears in.)
error: Command "gcc -pthread -fno-strict-aliasing -DNDEBUG -O2 -g
-pipe -Wp,-D_F
ORTIFY_SOURCE=3D2 -fexceptions -m32 -march=3Di386 -mtune=3Dpentium4
-fasynchronous-unw
ind-tables -D_GNU_SOURCE -fPIC -fPIC
-I/usr/lib/python2.4/site-packages/numpy/co
re/include -I/usr/local/include -I/usr/include
-I. -I/usr/include/python2.4 -c s
rc/_ns_cntr.c -o build/temp.linux-i686-2.4/src/_ns_cntr.o -DSCIPY=3D1"
failed with
 exit status 1

Stefan,
> Is this normal? If so, how do I get around the problem? I also
> noticed that, even without extents, the image gets scaled after
> plotting.
Try to set the "_autoscale" parameter of your current 'axes' to False. That 
way, you should avoid any inopportune rescaling. For the image, try to use 
aspect='auto'.
For example,
P.imshow(x,extent=(0,x.shape[1],x.shape[0],0))
P.gca().set_autoscale_on(False)
P.gca().set_aspect('auto')
Let us know if it works
P.

Stefan,
That certainly looks to me like a bug, but it is not obvious to me af=
ter=20
a quick look where the bug is (although I suspect it is very simple),=
=20
and I can't look at it more right now. If someone else doesn't chime=
 in=20
with a fix, you might want to file a bug report on sourceforge to mak=
e=20
sure it is not forgotten. Maybe I can take another look within the ne=
xt=20
few days.
What do you mean when you say "the image gets scaled after plotting"?
Eric
Stefan van der Walt wrote:
> Hi all,
>=20
> I have a script that reads in mouse-click coordinates from an image=
.
> I noticed that, with image extents specified, the axes flip wheneve=
r I
> plot to them.
>=20
> This snippet demonstrates the behaviour I see:
>=20
> # -- START --
>=20
> import pylab as P
> import numpy as N
>=20
> # Generate test pattern
> x =3D N.arange(100).reshape(-1,1) + N.zeros(100)
>=20
> def click(event):
> print "Mouse click at (%f,%f)" % (event.xdata,event.ydata)
> P.plot([event.xdata],[event.ydata],'o')
> P.draw()
>=20
> P.imshow(x,extent=3D(0,x.shape[1],x.shape[0],0))
> P.connect('button_press_event',click)
> P.show()
>=20
> # -- END --
>=20
>=20
> Is this normal? If so, how do I get around the problem? I also
> noticed that, even without extents, the image gets scaled after
> plotting.
>=20
> I'd appreciate any advice!
>=20
> Thanks for your time.
> St=E9fan
>=20
> -------------------------------------------------------------------=
------
> Take Surveys. Earn Cash. Influence the Future of IT
> Join SourceForge.net's Techsay panel and you'll get the chance to s=
hare your
> opinions on IT & business topics through brief surveys -- and earn =
cash
> http://www.techsay.com/default.php?page=3Djoin.php&p=3Dsourceforge&=
CID=3DDEVDEV
> _______________________________________________
> Matplotlib-users mailing list
> Mat...@li...
> https://lists.sourceforge.net/lists/listinfo/matplotlib-users

Bill,
You can see an example of how to do something similar by looking at the 
QuiverKey class in quiver.py. It is all a matter of using the 
transforms module.
Eric
Bill Baxter wrote:
> I want to draw some labels with plot.text() and have them appear a
> given number of pixels (or mm, or points) to above and to the right of
> the data points they are describing. Is there some way to specify a
> screen offset from a point in graph coordinates? Like a method of the
> axes that converts from screen to graph coords? Using graph
> coordinates means that when I zoom in really close to see some
> details, the text labels are way out in never-never land.
> 
> Thanks,
> --bb

Howdy PGM,
Thanks for the reply.
On 7/28/06, PGM <pgm...@gm...> wrote:
> Bill,
> Could you post an example of the problem ? I must admit I'm slower than usual
> to react today, and I'm not sure I understand what you mean.
> I tried that:
>
> plot([3,],[3,],'o')
> gca().text(3,3,'TEXT')
I'm plotting a mesh. A graph of vertexes and edges. I want to put
labels on the vertexes, but I want them off to the side of the vertex
a little bit so they don't overlap the vertex marker and all the
incident edges so much. With the commands you have above, the lower
left corner of the text and the marker overlap a little bit.
plot([3,],[3,],'o')
gca().text(3+little_offset,3+little_offset,'TEXT')
is what I want, except the mesh has some dense parts that I need to
zoom in to see clearly. But when I zoom in, that little_offset is
magnified just like everything else and the text ends up being a huge
distance from the vertex on the screen.
> and couldn't repdouce what you were mentioning. By default, the text method of
> an axes uses data coords, not graph coords. You want an offset to data
> coords, right ?
No, I want the offset to move things by a fixed distance in screen
space. I.e. 5 pixels on the final plot no matter the zoom level.
I did find one workaround just now.
plot([3,],[3,],'o')
gca().text(3,3,' TEXT\n')
That's good enough for me for now, but it seems like it would be
generally useful to be able to have a way to specify screen space
offsets from graph coords when drawing. Is there not such a thing?
--bb

"Michael Schwager" <hw...@cy...> writes:
> In fact, I'd like to control its x and y offset because it's a
> really large image and I only have letter or 11x17 size paper, so
> I'd like to print it out in panels.
Perhaps this is better done by postprocessing the ps file with a
suitable utility? For example,
http://www.tug.org/tex-archive/support/poster/poster.c
may do what you want.
-- 
Jouni

Hello,
Using 0.87.4, I'm getting this traceback with a date plot:
/usr/local/stow/matplotlib-0.87.4/lib/python2.4/site-packages/matplotlib/dates.py in _from_ordinalf(x, tz)
 154 dt = datetime.datetime.fromordinal(ix)
 155 remainder = x - ix
--> 156 hour, remainder = divmod(24*remainder, 1)
 157 minute, remainder = divmod(60*remainder, 1)
 158 second, remainder = divmod(60*remainder, 1)
ValueError: need more than 0 values to unpack
> /usr/local/stow/matplotlib-0.87.4/lib/python2.4/site-packages/matplotlib/dates.py(156)_from_ordinalf()
 155 remainder = x - ix
--> 156 hour, remainder = divmod(24*remainder, 1)
 157 minute, remainder = divmod(60*remainder, 1)
And sure enough,
 ipdb> divmod(0.0,1)
 (0.0, 0.0)
 ipdb> divmod( 24*remainder, 1 )
 ()
It looks like divmod is getting overloaded (?) or something in
CXX/cxx_extensions.cxx with [perhaps] this code:
 Py::Object PythonExtensionBase::number_divmod( const Py::Object & )
 { missing_method( number_divmod ); return Py::Nothing(); }
which would make sense, but I haven't explored it that thoroughly. I
think what's happening is that I've got a numpy array scalar that is
sneaking in there somewhere.
For now, I'll just use the hack-fix of casting remainder to a float.
Thanks,
Glen

Dear matplotlib-users,
I'd like to report a bug in Polygon, which is crashing with an 
unhelpful error message where an exception would be appropriate. The 
problem occurs when you feed Polygon an Nx2 array instead of an N- 
length list of 2-tuples. This is on my PPC OSX system, with 
everything freshly checked out from SVN (should the matplotlib 
version still be 0.87.4?).
Versions:
In [152]: numpy.__version__
Out[152]: '1.1.2881'
In [154]: matplotlib.__version__
Out[154]: '0.87.4'
Code:
import pylab, numpy
theta = numpy.pi/4*numpy.arange(9,dtype=float)
x = numpy.cos(theta)
y = numpy.sin(theta)
# The following line works
#p = pylab.Polygon(zip(x,y))
# The following line causes a crash
p = pylab.Polygon(numpy.vstack((x,y)).T)
ax = pylab.subplot(111)
ax.add_patch(p)
pylab.show()
Output:
In [155]: run plot_polygon.py
------------------------------------------------------------------------ 
---
exceptions.TypeError Traceback (most 
recent call last)
/Users/nvf/Documents/S.M. Thesis/plot_polygon.py
 10
 11 ax = pylab.subplot(111)
---> 12 ax.add_patch(p)
 13 pylab.show() 14
/Library/Frameworks/Python.framework/Versions/2.4/lib/python2.4/site- 
packages/matplotlib/axes.py in add_patch(self, p)
 899 p.get_transform(), p.get_verts())
 900 #for x,y in xys: print x,y
--> 901 self.update_datalim(xys)
 902 self.patches.append(p)
 903
/Library/Frameworks/Python.framework/Versions/2.4/lib/python2.4/site- 
packages/matplotlib/axes.py in update_datalim(self, xys)
 913 # Otherwise, it will compute the bounds of it's 
current data
 914 # and the data in xydata
--> 915 self.dataLim.update(xys, -1)
 916
 917
TypeError: CXX : Error creating object of type N2Py5TupleE
WARNING: Failure executing file: <plot_polygon.py>
Instead of converting from crash to exception, though, would it be 
possible to make it accept an Nx2 array?
Please at least cc me in any replies, as I am not subscribed to this 
list.
Thanks,
Nick

HI,
There are series like that...
> I need to plot in West
> Longitude, where the left edge of the graph starts at 360 and the right
> edge ends at 0. Does anyone know how to do this?
Assuming you have a plot (not an image):
xlim(360,0)
or gca.set_xlim(360,0)
HIH,
P.

Hi all,
I have a script that reads in mouse-click coordinates from an image.
I noticed that, with image extents specified, the axes flip whenever I
plot to them.
This snippet demonstrates the behaviour I see:
# -- START --
import pylab as P
import numpy as N
# Generate test pattern
x =3D N.arange(100).reshape(-1,1) + N.zeros(100)
def click(event):
 print "Mouse click at (%f,%f)" % (event.xdata,event.ydata)
 P.plot([event.xdata],[event.ydata],'o')
 P.draw()
P.imshow(x,extent=3D(0,x.shape[1],x.shape[0],0))
P.connect('button_press_event',click)
P.show()
# -- END --
Is this normal? If so, how do I get around the problem? I also
noticed that, even without extents, the image gets scaled after
plotting.
I'd appreciate any advice!
Thanks for your time.
St=E9fan

On Monday 24 July 2006 15:12, Jouni K Seppanen wrote:
> Tommy Grav <tg...@ma...> writes:
> > The window resizes horisontally when the pointer is inside the
> > window, although the window behaves erratically as it resizes.
> > Moving the pointer to the right will cause the window to grow then
> > reduce size as you go further right.
>
> Is this related to the pointer coordinates shown in the lower-right
> corner of the window? I.e., when you move the pointer right, perhaps
> the x-coordinate becomes a longer number and the window is enlarged to
> accommodate it. I remember seeing behavior like this on some old
> version of Matplotlib (probably TkAgg backend on OS X) but cannot
> reproduce it now.
This was the problem. I fixed it in the qt backends by setting the label's 
resize policy to ignore the sizing hints. 
(I'm sorry if this is the second time this message gets posted.)

I want to draw some labels with plot.text() and have them appear a
given number of pixels (or mm, or points) to above and to the right of
the data points they are describing. Is there some way to specify a
screen offset from a point in graph coordinates? Like a method of the
axes that converts from screen to graph coords? Using graph
coordinates means that when I zoom in really close to see some
details, the text labels are way out in never-never land.
Thanks,
--bb

Hi, I'm trying to print some stuff with a PS backend (python 2.4 with
WinXP). I have a figure, then I save it with savefig. Right now the figure
is centered in the page. However, if I change the page size (such as when
viewing with ghostscript), the figure is no longer centered. In fact, I'd
like to control its x and y offset because it's a really large image and I
only have letter or 11x17 size paper, so I'd like to print it out in panels.
How can I do this? Also, how can I specify 11x17 paper in savefig?
 
Thanks
Michael

Hello All,
I am trying to map the surface of TItan for a summer internship project at 
NASA. I would like to use matplotlib to plot, but I need to plot in West 
Longitude, where the left edge of the graph starts at 360 and the right edge 
ends at 0. Does anyone know how to do this?
Thanks much,
Jeff

Hi, I'm trying to do some CAD type things with Matplotlib. I'm using python
2.4 with both PS and GTK backends in WinXP. Specifically, I'm trying to
find distances between objects (lines, polygons, circles), so I'd like to
extract all the elements/lines/polygons/arcs/circles in an axis, but I don't
see how to do that. 
 
Also, I need to draw arcs, but it seems I can only do that by directly
manipulating the renderer. I'm trying to get everything done using only the
pylab interface. I tried creating a RendererPS directly, but it took some
funny arguments (eg a StreamIO) which I wasn't sure what to do with.
 
 
Any suggestions would be great.
Thanks
Michael

On Tuesday 25 July 2006 12:30, massimo sandal wrote:
> Hi,
>
> What is the method used to flip axes in X and/or Y direction in the
> matplotlib API? that is, to plot something with values from positive to
> negative instead of the contrary?
just set the limits of your axes with xlim/ylim or set_ylim/set_xlim
For example:
y = N.arange(10)
plot(y)
ylim(y[-1],y[0])

Hi,
What is the method used to flip axes in X and/or Y direction in the 
matplotlib API? that is, to plot something with values from positive to 
negative instead of the contrary?
Maybe it's me being dense :), but I can't find how to do it.
Thanks!
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

Charlie Moad ha scritto:
> I am guessing you installed numpy1.0b1. You need to use 0.9.8.
Thanks a lot. I uninstalled 1.01b1 and installed 0.9.8 and everything 
workes perfectly.
Thanks again,
Massimo
-- 
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

Charlie Moad wrote:
> Not really. I am using Visual Studio for the build in which I have
> never seen or heard of this error occuring. This was a mingw problem,
didn't the OP get this error with a MSVC build?
Anyway, I don't have windows booted to test myself, and I'm out of my depth.
-CHB
-- 
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...

Charlie Moad wrote:
> Well, I meant before this post. That is why I am somewhat skeptical of
> it. There have been hundreds of downloads, and this is the first
> message.
Hence my suspicion that this was a dll that is not included in Stock 
Windows, but is on most people's systems, installed by other software. I 
think the OP said s/he was a new (or infrequent) Windows user.
-CHB
-- 
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...

On Monday 24 July 2006 15:12, Jouni K Seppanen wrote:
> Tommy Grav <tg...@ma...> writes:
> > The window resizes horisontally when the pointer is inside the
> > window, although the window behaves erratically as it resizes.
> > Moving the pointer to the right will cause the window to grow then
> > reduce size as you go further right.
>
> Is this related to the pointer coordinates shown in the lower-right
> corner of the window? I.e., when you move the pointer right, perhaps
> the x-coordinate becomes a longer number and the window is enlarged to
> accommodate it. I remember seeing behavior like this on some old
> version of Matplotlib (probably TkAgg backend on OS X) but cannot
> reproduce it now.
That looks like the problem. If you click the zoom widget in qt4agg, 
then "zoom to rect mode" also appears on that line, forcing the width of the 
window to increase even further.

Tommy Grav <tg...@ma...> writes:
> The window resizes horisontally when the pointer is inside the
> window, although the window behaves erratically as it resizes. 
> Moving the pointer to the right will cause the window to grow then
> reduce size as you go further right.
Is this related to the pointer coordinates shown in the lower-right
corner of the window? I.e., when you move the pointer right, perhaps
the x-coordinate becomes a longer number and the window is enlarged to
accommodate it. I remember seeing behavior like this on some old
version of Matplotlib (probably TkAgg backend on OS X) but cannot
reproduce it now.
-- 
Jouni

On 7/24/06, Chris Fonnesbeck <fon...@gm...> wrote:
> And by the way, trying to build matplotlib myself with the numpy
> binary installed, fails. It appears to be related to tkagg:
>
> src/_tkagg.cpp:28:18: tk.h: No such file or directory
> src/_tkagg.cpp:36: syntax error before `*' token
> src/_tkagg.cpp:40: `ClientData' was not declared in this scope
> src/_tkagg.cpp:40: parse error before `,' token
> src/_tkagg.cpp: In function `int PyAggImagePhoto(...)':
> src/_tkagg.cpp:43: `Tk_PhotoHandle' undeclared (first use this function)
> src/_tkagg.cpp:43: (Each undeclared identifier is reported only once for each
> function it appears in.)
> src/_tkagg.cpp:43: parse error before `;' token
> src/_tkagg.cpp:44: `Tk_PhotoImageBlock' undeclared (first use this function)
> src/_tkagg.cpp:56: `argc' undeclared (first use this function)
> src/_tkagg.cpp:57: `interp' undeclared (first use this function)
> src/_tkagg.cpp:57: `argv' undeclared (first use this function)
> src/_tkagg.cpp:58: `Tcl_AppendResult' undeclared (first use this function)
> src/_tkagg.cpp:59: `TCL_ERROR' undeclared (first use this function)
> src/_tkagg.cpp:63: `photo' undeclared (first use this function)
> src/_tkagg.cpp:63: `Tk_FindPhoto' undeclared (first use this function)
> src/_tkagg.cpp:117: `block' undeclared (first use this function)
> src/_tkagg.cpp:143: `Tk_PhotoPutBlock' undeclared (first use this function)
> src/_tkagg.cpp:153: `Tk_PhotoBlank' undeclared (first use this function)
> src/_tkagg.cpp:158: `TCL_OK' undeclared (first use this function)
> src/_tkagg.cpp: In function `PyObject* _tkinit(PyObject*, PyObject*)':
> src/_tkagg.cpp:174: `Tcl_Interp' undeclared (first use this function)
> src/_tkagg.cpp:183: parse error before `)' token
> src/_tkagg.cpp:188: `struct TkappObject' has no member named `interp'
> src/_tkagg.cpp:194: `Tcl_CmdProc' undeclared (first use this function)
> src/_tkagg.cpp:194: parse error before `)' token
> C:/MinGW/include/c++/3.2.3/bits/basic_string.tcc: At top level:
> src/_tkagg.cpp:42: warning: `int PyAggImagePhoto(...)' defined but not used
> error: Command "g++ -O2 -Wall -Iwin32_static/include/tcl84 -I. -Isrc -Iswig -Iag
> g23/include -I. -I. -Iwin32_static/include/tcl84\freetype2 -I.\freetype2 -Isrc\f
> reetype2 -Iswig\freetype2 -Iagg23/include\freetype2 -I.\freetype2 -I.\freetype2
> -IC:\Python24\lib\site-packages\numpy\core\include -IC:\Python24\include -IC:\Py
> thon24\PC -c src/_tkagg.cpp -o build\temp.win32-2.4\Release\src\_tkagg.o" failed
> with exit status 1
>
> However, I have Tk installed, as it comes bundled with ActiveState python.
are you running:
win32_static\profile24.bat
win32_static\importlib24.bat

>>>>> "Chris" == Chris Fonnesbeck <fon...@gm...> writes:
 Chris> And by the way, trying to build matplotlib myself with the
 Chris> numpy binary installed, fails. It appears to be related to
 Chris> tkagg:
 Chris> src/_tkagg.cpp:28:18: tk.h: No such file or directory
 ...snip...
 Chris> However, I have Tk installed, as it comes bundled with
 Chris> ActiveState python.
You need the tk devel headers -- do you have tk.h? Is it in your
include path?
JDH