matplotlib-users Mailing List for matplotlib

Jeff Sadino wrote:
> I am trying to map the surface of Saturn's largest moon, Titan for a summer research project. As a rough draft, I would like to plot the temperature (colored red) at the longitude(x) and latitude(y) coordinates. My data file is tab delimited in three columns (x,y,T). I am also missing some x,y locations, which I would like to leave blank.
>
> As a final report, I would like the data point to be able to take on different geometries (circle, hexagon, arbitrary shape, etc) and different sizes as well as be able to overlap with neighboring data points. Finally, I would like to be able to plot this not on a rectangular cartesion coordinate system, but on a planetographic (global projection) coordinate system.
>
> I've spent two days trying everything from gnuplot, matlab, matplotlib and others trying to do this, but to no avail. Any suggestions would be very welcomed.
>
>
> 
Jeff: The Basemap toolkit can plot data on may different map 
projections. Here's an example that plots some random data on an 
orthographic projection:
from matplotlib.toolkits.basemap import Basemap
import pylab as p
from matplotlib.numerix.random_array import uniform
# set up orthographic map projection with
# perspective of satellite looking down at 50N, 100W.
# use resolution = None, since you don't want earth geography.
map = Basemap(projection='ortho',lat_0=50,lon_0=-100,resolution=None)
# draw the edge of the map projection region (the projection limb)
map.drawmapboundary()
# draw lat/lon grid lines every 30 degrees.
map.drawmeridians(p.arange(0,360,30))
map.drawparallels(p.arange(-90,90,30))
# number of points to plot.
npts = 750
# generate random points on a sphere,
# so that every small area on the sphere is expected
# to have the same number of points.
# http://mathworld.wolfram.com/SpherePointPicking.html
u = uniform(0.,1.,size=npts) # this is for numpy, for Numeric use 
shape=npts
v = uniform(0.,1.,size=npts)
lons = 360.*u
lats = (180./p.pi)*p.arccos(2*v-1) - 90.
# transform lons and lats to map coordinates.
x,y = map(lons,lats)
# random intensities between 0 and 1
z = uniform(0.,1.,size=npts)
# plot filled circles at the locations of the points
# with size=202 points, with intensities colored using jet colormap.
# you can change the shape of the marker with the 'marker' keyword
# faceted=False makes the edge colors the same as the fill colors
# see http://matplotlib.sourceforge.net/matplotlib.pylab.html#-scatter
map.scatter(x,y,20,z,cmap=p.cm.jet,marker='o',faceted=False)
p.show()
-Jeff
-- 
Jeffrey S. Whitaker Phone : (303)497-6313
Meteorologist FAX : (303)497-6449
NOAA/OAR/PSD R/PSD1 Email : Jef...@no...
325 Broadway Office : Skaggs Research Cntr 1D-124
Boulder, CO, USA 80303-3328 Web : http://tinyurl.com/5telg

Darren Dale wrote:
> On Wednesday 12 July 2006 16:16, Mark Bakker wrote:
> 
>>I am following up on the discussion of passing a single 2D array to plot.
>>Wouldn't it make more sense that, in Python array style,
>>if you give it a single N x K argument you plot rows against the first row?
> 
> 
> That's not the behavior I would have expected. I would have expected each row 
> to be plotted as a funtion of the column index, just like plot([1,2,3,4]) is 
> done now, but with each row in a different color. I would like 
> plot([1,2,3,4], array([[1,2,3,4],[2,3,4,5]])) to plot two lines as a function 
> of the single x-list, and I would like plot(array([[1,2,3,4],[2,3,4,5]]), 
> array([[1,2,3,4],[2,3,4,5]])) to take the next obvious step. I *think* matlab 
> does this, but its been so long since I used it...
I was afraid someone would bring up these possibilities...
One of the problems with fancy argument handling is that there are many 
ways it can be done, so we have to decide on the behavior we want; and 
inevitably this will surprise (and probably irritate) some people just 
as it pleases others. (Other problems include complex code, complex 
documentation, and runtime overhead. For interactive use, however, 
fancy argument handling may be worth the trouble.)
We are going to have some tradeoffs among matlab compatibility, internal 
consistency, and adherence to the underlying data storage model.
Here is what Matlab does (and in one respect it is not what I thought I 
remembered):
1) Given a single NxK matrix, it plots each column against the row index.
2) Given a vector and an NxK, it plots each column against the vector.
3) Given an NxK and a vector, it plots the vector against each column.
4) Given an NxK and an NxK it plots each column from the second against 
the corresponding column of the first.
I think this is a good model: fairly simple, consistent, intuitive, and 
covers a good range of real-life situations. It differs from what I 
thought I remembered, and from what I think Stefan requested, in that, 
given an Nx2, it does not plot the second column against the first. 
That behavior, however, does not generalize nicely to NxK for any K!=2, 
so I now think we should choose either one or the other. A virtue of 
Stefan's Nx2 proposal is that it is consistent with the changes I made 
to elsewhere so that paths can be specified as Nx2 arrays; this, in 
turn, was consistent with the original specification as sequences of 
(x,y) tuples. But plot has never used sequences of (x,y) tuples, so the 
argument for the Nx2 form is weaker here.
I don't know whether the reason Matlab chooses the columns as the data 
vectors is because of the Fortran storage order Matlab uses, or whether 
there is some other reason. Personally I am very comfortable with it, 
perhaps simply because of my Matlab experience. I think part of it is 
that columns in a table seem more natural as data vectors than rows, 
however; tables are usually oriented so that columns (fields) are 
different variables, and the row index is the sample number, or time, or 
a spatial coordinate.
To summarize, the options seem to be:
1) Leave plot argument parsing alone.
2) Accept an Nx2 array in place of a pair of arguments containing x and y.
3) Implement the Matlab model.
4) Implement the Matlab model, but taking rows instead of columns in an 
X or Y array that is 2-D.
I am open to arguments, but my preference is the Matlab model. I don't 
think that the difference in native array storage order matters much. 
It is more important to have the API at the plot method and function 
level match the way people think.
Eric

matplotlib+basemap may do it for you. I'm not sure how much of it's hard 
coded for the Earth though...
http://www.scipy.org/Cookbook/Matplotlib/Maps
At 08:49 AM 7/12/2006, Jeff Sadino wrote:
>I am trying to map the surface of Saturn's largest moon, Titan for a 
>summer research project. As a rough draft, I would like to plot the 
>temperature (colored red) at the longitude(x) and latitude(y) 
>coordinates. My data file is tab delimited in three columns (x,y,T). I 
>am also missing some x,y locations, which I would like to leave blank.
>
>As a final report, I would like the data point to be able to take on 
>different geometries (circle, hexagon, arbitrary shape, etc) and different 
>sizes as well as be able to overlap with neighboring data 
>points. Finally, I would like to be able to plot this not on a 
>rectangular cartesion coordinate system, but on a planetographic (global 
>projection) coordinate system.
>
>I've spent two days trying everything from gnuplot, matlab, matplotlib and 
>others trying to do this, but to no avail. Any suggestions would be very 
>welcomed.
>
>
>
>-------------------------------------------------------------------------
>Using Tomcat but need to do more? Need to support web services, security?
>Get stuff done quickly with pre-integrated technology to make your job easier
>Download IBM WebSphere Application Server v.1.0.1 based on Apache Geronimo
>http://sel.as-us.falkag.net/sel?cmd=lnk&kid=120709&bid=263057&dat=121642
>_______________________________________________
>Matplotlib-users mailing list
>Mat...@li...
>https://lists.sourceforge.net/lists/listinfo/matplotlib-users
Ted Drain Jet Propulsion Laboratory ted...@jp... 

Mark Bakker wrote:
> 
> On the same token, it would be really nice if contour, pcolor and image
> take as an x and y argument not only a matrix but just a 1D row. This
> should be really easy and would be very useful.
contour already does this, and I agree that pcolor and pcolormesh 
should. pcolor* is a little more complicated than contour because the X 
and Y dimensions are ideally larger by 1 than the corresponding Z 
dimensions, but don't have to be.
image does not take x,y arguments at all, by design.
> 
> In fact, if more people want it I can submit a patch for the latter one,
OK. I suggest a function that can be used for both pcolor and 
pcolormesh. (Some day the two should be consolidated, but there is a 
pcolormesh bug with alpha != 1, so we can't do it yet.)
Eric

Hi:
I obtain the following error when saving a figure to postscript after
running one of the test routines in mplot3d.py:
>>> from numpy import *; import mpl3d.mplot3d as p3; import pylab as p
>>> p3.test1()
>>> p.savefig('test1')
Traceback (most recent call last):
 File "<stdin>", line 1, in ?
 File "/usr/local/lib/python2.4/site-packages/matplotlib/pylab.py",
line 811, in savefig
 return fig.savefig(*args, **kwargs)
 File "/usr/local/lib/python2.4/site-packages/matplotlib/figure.py",
line 660, in savefig
 self.canvas.print_figure(*args, **kwargs)
 File "/usr/local/lib/python2.4/site-packages/matplotlib/backends/backend_ps.py",
line 1061, in print_figure
 self.figure.draw(renderer)
 File "/usr/local/lib/python2.4/site-packages/matplotlib/figure.py",
line 531, in draw
 for a in self.axes: a.draw(renderer)
 File "/usr/local/lib/python2.4/site-packages/mpl3d/mplot3d.py", line
714, in draw
 self.w_xaxis.draw(renderer)
 File "/usr/local/lib/python2.4/site-packages/mpl3d/mplot3d.py", line
613, in draw
 tick.draw(renderer)
 File "/usr/local/lib/python2.4/site-packages/matplotlib/axis.py",
line 161, in draw
 if self.label1On: self.label1.draw(renderer)
 File "/usr/local/lib/python2.4/site-packages/matplotlib/text.py",
line 1166, in draw
 self.update_coords(renderer)
 File "/usr/local/lib/python2.4/site-packages/mpl3d/mplot3d.py", line
411, in update_coords
 return text_update_coords(self, renderer)
 File "/usr/local/lib/python2.4/site-packages/mpl3d/mplot3d.py", line
102, in text_update_coords
 we = self._mytext.get_window_extent(renderer=renderer)
AttributeError: TextWithDash instance has no attribute '_mytext'
I installed matplotlib-0.87.3 and mpl3d yesterday and today. After a
bit of hunting, I noticed there are two similar definitions of
TextWithDash in matplotlib's text.py: _TextWithDash and TextWithDash.
Noting the underscored version defines _mytext but the underscore-free
version does not, I switched the underscores. Now things are working
(though perhaps I've broken something else in the process). I suspect
the error is a product of on-going changes to matplotlib; nonetheless,
I thought my email might be helpful to others.
Andrew

On Wednesday 12 July 2006 16:16, Mark Bakker wrote:
> I am following up on the discussion of passing a single 2D array to plot.
> Wouldn't it make more sense that, in Python array style,
> if you give it a single N x K argument you plot rows against the first row?
That's not the behavior I would have expected. I would have expected each row 
to be plotted as a funtion of the column index, just like plot([1,2,3,4]) is 
done now, but with each row in a different color. I would like 
plot([1,2,3,4], array([[1,2,3,4],[2,3,4,5]])) to plot two lines as a function 
of the single x-list, and I would like plot(array([[1,2,3,4],[2,3,4,5]]), 
array([[1,2,3,4],[2,3,4,5]])) to take the next obvious step. I *think* matlab 
does this, but its been so long since I used it...
Darren

On Wednesday 12 July 2006 09:49, Darren Dale wrote:
> Hello, um..., P.,
>
> On Wednesday 12 July 2006 03:43, PGM wrote:
> > Folks,
> > I need your wisdom about ticks labels on ordinates for large numbers
> > (>1e4). The default behavior I have (0.87.4) is to display tick labels as
> > "%.1f", and write a string "x1e+..." above the top left corner of the
> > current axes.
> >
> > - When using "yaxis.tick_right()", the "x1e..." string stays above the
> > top left corner. That becomes an issue when using two different scales.
> > How could I force the 'mantissa' string to be on the same side as the
> > axis it depends on ?
>
> I'll take care of this, but I will need a few days.
As of svn 2560, the label will render over the left or right y-axis depending 
on the position of the ticks.
Darren

I am a newbie to python, numpy, and matplotlib, and I like what I have 
seen so far. I have been wishing for a long time there was an open 
alternative to matlab, and when I heard about numpy and matplotlib, I 
tried them out immediately.
Anyway, I am running python embedded in a C++ WX app. I just discovered 
matplotlib, so I tried it out. From the interactive python shell 
(vanilla FC5 linux install of python and matplotlib) everything works 
fine. The show() command blocks.
In my application, I use the PyRun_SimpleFile command to execute user 
selected python files on a separate thread. That has seemed to work 
fine for several weeks on scripts using numpy and such. When I use 
matplotlib, however, the show() command does not block, and then strange 
things happen when I try to close the Figure window. On the first close 
it reopens the window right back up. On the second close attempt it 
closes (gdb splits out zombie thread messages), and then if I try to 
execute another script with a show() command in it it sometimes seg 
faults, sometimes works, but in general causes my program to behave 
strangely.
So my question is: what is the correct way to setup python and/or 
matplotlib when running python embedded? Is there a way to make show() 
blocking when using embedded python (I am not sure this will solve any 
problems).
Thanks,
Lane Brooks

I am following up on the discussion of passing a single 2D array to plot.
Wouldn't it make more sense that, in Python array style,
if you give it a single N x K argument you plot rows against the first row?
On the same token, it would be really nice if contour, pcolor and image
take as an x and y argument not only a matrix but just a 1D row. This
should be really easy and would be very useful.
In fact, if more people want it I can submit a patch for the latter one,
Mark
------------------------------
>
> Message: 4
> Date: 2006年7月12日 14:38:46 -0400
> From: Alan G Isaac <ai...@am...>
> Subject: Re: [Matplotlib-users] ANN: matplotlib-0.87.4 (bugfix release
> for enthon)
> To: mat...@li...
> Message-ID: <Mah...@am...>
> Content-Type: TEXT/PLAIN; CHARSET=ISO-8859-1
>
> On 2006年7月12日, Eric Firing apparently wrote:
> > In the single-argument NxK case, I think matlab plots
> > subsequent columns against the first column. Is this what
> > you would like?
>
> Yes.
>
> Cheers,
> Alan Isaac
>
>

>>>>> "Gregory" =3D=3D Gregory Pi=F1ero <gre...@gm...> writes:
 Gregory> If correct here, I guess I locate the API in the user
 Gregory> guide? Is there anything else I should be using as a
 Gregory> reference?
http://matplotlib.sourceforge.net/faq.html#OO
http://matplotlib.sourceforge.net/leftwich_tut.txt
JDH

On 7/12/06, John Hunter <jdh...@ac...> wrote:
> In a production app like a web application server generating graphics,
> often times you don't want any magic going on behind the hood, and
> pylab's stateful management of figures, the current figure, and the
> current axes grates on some people who like to have full control over
> object creation and destruction.
>
> So the webapp demo shows how to use the API to manage and use
> figures. It is mainly a style choice.
Thanks, I'm starting to understand. So since I am making a web app, I
should be refering to the API reference instead of the screenshots
code:
http://matplotlib.sourceforge.net/screenshots.html
If correct here, I guess I locate the API in the user guide? Is there
anything else I should be using as a reference?
-Greg

On 7/12/06, John Hunter <jdh...@ac...> wrote:
>
> >>>>> "John" == John Hunter <jdh...@ac...> writes:
>
> >>>>> "David" == David Grant <dav...@gm...> writes:
> David> edges=[54, 76, 80, 100] stats=[5.423, 23.226, 4.1, 6.93]
>
>
> John> xticks(edges)
>
> Sorry, what I meant was something like
>
> xticks(range(len(edges)), ['%d'%edge for edge in edges])
>
> This will set the locations to the integers and the labels to your
> edge labels.
Cool, yeah I think that will work just fine. I don't think I saw xticks in
the matplotlib PDF document I was looking at. What's the best source of
documentation? I just noticed it is here:
http://matplotlib.sourceforge.net/pylab_commands.html which is where I
looked originally... I guess I just didn't know what to look for. :-)
p.s. any idea why [matplotlib-users] didn't show up in the subject line of
my post?
-- 
David Grant
Please Note my new email address: dav...@gm...

>>>>> "David" == David Grant <dav...@gm...> writes:
 David> Cool, yeah I think that will work just fine. I don't think
 David> I saw xticks in the matplotlib PDF document I was looking
 David> at. What's the best source of documentation? 
help(pylab)
JDH

>>>>> "Gregory" =3D=3D Gregory Pi=F1ero <gre...@gm...> writes:
 Gregory> So my only remaining point of confusion is why wasn't the
 Gregory> pylab interface used in the demo here:
 Gregory> http://matplotlib.sourceforge.net/examples/webapp_demo.py
 Gregory> Is it just a style choice?
pylab manages figures for you -- you want to be sure in an application
that creates many figures and saves them to images (like a web app)
that every figure that is created is closed
 fig =3D figure(1)
 ...later
 close(1)
In a production app like a web application server generating graphics,
often times you don't want any magic going on behind the hood, and
pylab's stateful management of figures, the current figure, and the
current axes grates on some people who like to have full control over
object creation and destruction.
So the webapp demo shows how to use the API to manage and use
figures. It is mainly a style choice.
JDH

So my only remaining point of confusion is why wasn't the pylab
interface used in the demo here:
> Gregory> http://matplotlib.sourceforge.net/examples/webapp_demo.py
Is it just a style choice?

>>>>> "Gregory" =3D=3D Gregory Pi=F1ero <gre...@gm...> writes:
 Gregory> Hi guys, Is this possible? Specifically I'm trying
 Gregory> implement something close to this example:
 Gregory> http://matplotlib.sourceforge.net/examples/webapp_demo.py
 Gregory> But would like to be able to utilize code found in
 Gregory> samples like this:
 Gregory> http://matplotlib.sourceforge.net/screenshots/pie_demo.py
You just have to tell matplotlib to use Agg as your backend -- pylab
will respect the default backend, whether it is a GUI, Agg or PS.
There are several ways to choose the backend
 * make it permanent by setting your backend to 'Agg' in matplotlibrc
 * change it on a per script basis exterally by launching with the -d
 option
 > python myscript.py -dAgg
 > python myscript.py -dPS
 > python myscript.py -dGTKAgg
 * probably best for a web app, use the matplotlib "use directive".
 This must be done *before* importing pylab
 import matplotlib
 matplotlib.use('Agg')
 import pylab
 plot and save....
This is discussed at http://matplotlib.sf.net/backends.html
JDH

Hi guys,
Is this possible? Specifically I'm trying implement something close
to this example:
http://matplotlib.sourceforge.net/examples/webapp_demo.py
But would like to be able to utilize code found in samples like this:
http://matplotlib.sourceforge.net/screenshots/pie_demo.py
Much thanks!
--=20
Gregory Pi=F1ero
Chief Innovation Officer
Blended Technologies
(www.blendedtechnologies.com)

>>>>> "John" == John Hunter <jdh...@ac...> writes:
>>>>> "David" == David Grant <dav...@gm...> writes:
 David> edges=[54, 76, 80, 100] stats=[5.423, 23.226, 4.1, 6.93]
 John> xticks(edges)
Sorry, what I meant was something like
 xticks(range(len(edges)), ['%d'%edge for edge in edges]) 
This will set the locations to the integers and the labels to your
edge labels.
JDH

>>>>> "David" == David Grant <dav...@gm...> writes:
 David> edges=[54, 76, 80, 100] stats=[5.423, 23.226, 4.1, 6.93]
xticks(edges) 
perhaps?
Or maybe I still don't understand.
A complete script might help...
JDH

On 7/12/06, John Hunter <jdh...@ac...> wrote:
>
> >>>>> "David" == David Grant <dav...@gm...> writes:
>
> David> Anyone know what happened to
> David> matplotlib.ticker.IndexFormatter? Is there are replacement
> David> for it?
>
> I don't know -- I don't see anything in the changelog or in the svn
> log. What did it do, and what version of mpl had it? Sorry, but my
> memory is failing here... I don't see any IndexFormatter as far back
> as 0.83. There is an index locator however....
I managed to accomplish what I wanted to do, using the FuncFormatter I
think... Here's what I was doing. I am plotting results of graphs, graphs
which have a certain number of edges. So I wanted to plot some statistic vs.
number of edges. But the number of edges was not varied in a linear way but
I have a list of which edges. So:
edges=[54, 76, 80, 100]
stats=[5.423, 23.226, 4.1, 6.93]
So basically I wanted t=arange(len(stats) as my x-axis but edges for the
labels (but only if it lies on a major tic). So I did:
 majorFormatter = ticker.FuncFormatter(lambda x, pos: labels[int(x)])
 axis.xaxis.set_major_formatter(majorFormatter)
If anyone knows a better way I would love to know. I'm just learning
matplotlib (I'm a former matlab user) and would love to master it.
-- 
David Grant
Please Note my new email address: dav...@gm...

On 2006年7月12日, Eric Firing apparently wrote:=20
> In the single-argument NxK case, I think matlab plots=20
> subsequent columns against the first column. Is this what=20
> you would like?=20
Yes.
Cheers,
Alan Isaac

Alan G Isaac wrote:
> On 2006年7月12日, Stefan van der Walt apparently wrote:=20
>=20
>>It would be useful to have plot accept a 2-D array as well. Would=
=20
>>patches for this be considered, or is there some reason why this ca=
n't=20
>>work? At the moment, doing=20
>>P.plot(z) where z is Nx2=20
>=20
>=20
> If 2-D is allowed,
> I hope N=D7K is allowed,
> not just N=D72.
I have thought about this, but haven't gotten around to doing it. To=
 do=20
the whole job, we could handle a single NxK argument, and we could al=
so=20
handle pairs:
Nx1, NxK
NxK, Nx1
NxK, NxK
I think it is clear what the behavior should be with any of the pairs=
;=20
essentially they should be broadcast to the NxK, NxK case, resulting =
in=20
K lines.
In the single-argument NxK case, I think matlab plots subsequent colu=
mns=20
against the first column. Is this what you would like? The alternat=
ive=20
would be to plot each column against the row index, which is what=
=20
happens in the Nx1 case, but I suspect the matlab-compatibility argum=
ent=20
wins out here. Also, this maintains consistency with the original=
=20
request for Nx2, which I assume was for plotting the second column=
=20
against the first. I certainly would not want to have NxK behavior=
=20
change as K goes from 2 to values larger than 2.
Eric

I am trying to map the surface of Saturn's largest moon, Titan for a summer research project. As a rough draft, I would like to plot the temperature (colored red) at the longitude(x) and latitude(y) coordinates. My data file is tab delimited in three columns (x,y,T). I am also missing some x,y locations, which I would like to leave blank.
As a final report, I would like the data point to be able to take on different geometries (circle, hexagon, arbitrary shape, etc) and different sizes as well as be able to overlap with neighboring data points. Finally, I would like to be able to plot this not on a rectangular cartesion coordinate system, but on a planetographic (global projection) coordinate system.
I've spent two days trying everything from gnuplot, matlab, matplotlib and others trying to do this, but to no avail. Any suggestions would be very welcomed.

>>>>> "David" == David Grant <dav...@gm...> writes:
 David> Anyone know what happened to
 David> matplotlib.ticker.IndexFormatter? Is there are replacement
 David> for it?
I don't know -- I don't see anything in the changelog or in the svn
log. What did it do, and what version of mpl had it? Sorry, but my
memory is failing here... I don't see any IndexFormatter as far back
as 0.83. There is an index locator however....
Perhaps you should just tell us what you want to do.
JDH

Anyone know what happened to matplotlib.ticker.IndexFormatter? Is there are
replacement for it?
-- 
David Grant