matplotlib-users Mailing List for matplotlib

Carol Leger wrote:
> What is 3D?  I see it in a lot of these posts, but I do not see it in
> any or the examples.
> 
> In reply to my question about color in quiver, it was suggested that I try:
> 
> quiver(x,y,u,v,color =3D (v>0)*1.) to draw arrows with positive v to the
> color corresponding to index 1 in the color map, and the arrows with
> negative v to the color corresponding to 0.

Some mail programs encode the text in a funny way that uses = as an escape
character. The encoded representation of a real = character is =3D.

http://en.wikipedia.org/wiki/Quoted-printable

-- 
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

What is 3D?  I see it in a lot of these posts, but I do not see it in 
any or the examples.

In reply to my question about color in quiver, it was suggested that I try:

quiver(x,y,u,v,color =3D (v>0)*1.) to draw arrows with positive v to the 
color corresponding to index 1 in the color map, and the arrows with 
negative v to the color corresponding to 0.
-- 
Ms. Carol A. Leger
SRI International			Phone: (650) 859-4114
333 Ravenswood Avenue G-273
Menlo Park, CA 94025                    e-mail: le...@sr...

Hi Carol,

For the first question, you could do something like
quiver(x,y,u,v,color =3D (v>0)*1.)
which would draw the arrows with positive v to the color corresponding
to index 1 in the colormap, and the arrows with negative v to the
color corresponding to 0. For the jet colormap, these are red and blue
respectively.

If you don't like red and blue, you could change the colors by
changing the colormap.

I tried using the "color array" to tell explicitely quiver what color
to use but I haven't been able to work it out.


Cheers,

David


2006/3/15, Carol Leger <car...@sr...>:
> Hi Folks,
>
> It looks like quiver might be able to what I want.
>
> However, I have a few more questions. Assume the form
> QUIVER( X, Y, U, V, S)
> 1)
> I see from the on-line documentation that:
>
> "color can be an array of colors in which case the arrows can be
> colored according to another dataset."
>
> What is an array of colors?  I want to make vectors one color if an
> element of U is negative and another color if it is zero or positive.
>
> 2)
> I want to make several plots, i.e a series of png or ps files, and have
> a vector with magnitude N be the same size on each plot. Is this what
> setting S equal to zero is supposed to accomplish?
>
>
> --
> Ms. Carol A. Leger
> SRI International                       Phone: (650) 859-4114
> 333 Ravenswood Avenue G-273
> Menlo Park, CA 94025                    e-mail: le...@sr...
>
>
> -------------------------------------------------------
> This SF.Net email is sponsored by xPML, a groundbreaking scripting langua=
ge
> that extends applications into web and mobile media. Attend the live webc=
ast
> and join the prime developer group breaking into this new coding territor=
y!
> http://sel.as-us.falkag.net/sel?cmd=3Dlnk&kid=3D110944&bid=3D241720&dat=
=3D121642
> _______________________________________________
> Matplotlib-users mailing list
> Mat...@li...
> https://lists.sourceforge.net/lists/listinfo/matplotlib-users
>

Hi Folks,

It looks like quiver might be able to what I want.

However, I have a few more questions. Assume the form
QUIVER( X, Y, U, V, S)
1)
I see from the on-line documentation that:

"color can be an array of colors in which case the arrows can be
colored according to another dataset."

What is an array of colors?  I want to make vectors one color if an 
element of U is negative and another color if it is zero or positive.

2)
I want to make several plots, i.e a series of png or ps files, and have 
a vector with magnitude N be the same size on each plot. Is this what 
setting S equal to zero is supposed to accomplish?


-- 
Ms. Carol A. Leger
SRI International			Phone: (650) 859-4114
333 Ravenswood Avenue G-273
Menlo Park, CA 94025                    e-mail: le...@sr...

I want to make a legend. Looking at the existing Legend class, I'm not 
sure I have precise control over positioning, which I would like to have 
(i.e., put it in a specific rectangle). I would also like to put legend 
items in a horizontal layout, or perhaps put them in columns.

I'm not sure Legend can do this. However, it is easy enough to make a 
do-it-yourself legend by creating an Axes and plotting points and text 
on it. Will this be necessary?

Mike

Thanks!

By pairing every figure command with a close, I avoid memory leaks and solv=
e
my problem.

On 3/15/06, John Hunter <jdh...@ac...> wrote:
>
> >>>>> "Alan" =3D=3D Alan Isaac <ai...@am...> writes:
>
>     Alan> On Wed, 15 Mar 2006, Imara Jarrett wrote:
>     >> I would like to generate matplotlib scatter plots in a python
>     >> 'for' loop.
>
>     Alan> Use 'figure'.
>
> You have to be a little careful here -- remember whenever you are
> creating plots in a loop you should pair every figure command with a
> close to avoid memory leaks
>
> for i in somerange:
>     figure(1)
>     scatter(mydata, ...)
>     savefig(myfile)
>     close(1)
>
> You have a few other alternatives....  The hold state affects whether
> new plot commands to an Axes/Subplot will overwrite the existing one
> or add on to it
>
>   scatter(mydata, ..., hold=3DFalse)  # clear the existing axes
>   scatter(mydata, ..., hold=3DTrue)   # add to the existing axes
>
> See also the commands clf (clear current figure) and cla (clear
> current axes)
>
> JDH
>
>
>
>
> -------------------------------------------------------
> 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=3Dlnk&kid=3D110944&bid=3D241720&dat=
=3D121642
> _______________________________________________
> Matplotlib-users mailing list
> Mat...@li...
> https://lists.sourceforge.net/lists/listinfo/matplotlib-users
>

>>>>> "Alan" == Alan Isaac <ai...@am...> writes:

    Alan> On Wed, 15 Mar 2006, Imara Jarrett wrote:
    >> I would like to generate matplotlib scatter plots in a python
    >> 'for' loop.

    Alan> Use 'figure'.

You have to be a little careful here -- remember whenever you are
creating plots in a loop you should pair every figure command with a
close to avoid memory leaks

for i in somerange:
    figure(1)
    scatter(mydata, ...)
    savefig(myfile)
    close(1)

You have a few other alternatives....  The hold state affects whether
new plot commands to an Axes/Subplot will overwrite the existing one
or add on to it

  scatter(mydata, ..., hold=False)  # clear the existing axes
  scatter(mydata, ..., hold=True)   # add to the existing axes

See also the commands clf (clear current figure) and cla (clear
current axes)

JDH

On Wed, 15 Mar 2006, Imara Jarrett wrote: 
> I would like to generate matplotlib scatter plots in a python 'for' loop. 

Use 'figure'.

hth,
Alan Isaac

Hi folks,

I would like to generate matplotlib scatter plots in a python 'for' loop.

However, I am running into the following problem.  The first scatter plot i=
s
fine, each subsequent scatter plot still has the points from the previous
scatter plot, even though I am generating a new list with each iteration.
This results in the last scatter plot to have a gigantic number of points.

Here is sample code:

for x in dates:
    xpoints =3D [ ]
    ypoints =3D [ ]
    size =3D [ ]
    . . . .
    . . . . .

    xticker =3D arange(0,6000, 1000)
    yticker =3D arange(-4, 10, 1)
    xticks(xticker)
    yticks(yticker)

    scatter(xpoints,ypoints, s =3D size)

    savefig('example.png')

Any suggestions?

Thanks in advance.

>>>>> "Nils" == Nils Wagner <nw...@me...> writes:

    >>>> matplotlib.__version__
    Nils> '0.87.2svn'

This question is really for the matplotlib list.  Please try to avoid
cross posting.  There is a lot of overlap between the readers of both
lists and you can be pretty sure that most of the mpl and scipy
developers will see it in either place.

What is your numerix setting and what is it's version?  You can obtain
this by creating a little script that just does 'import pylab'

and then running it with

> python myscript.py --verbose-helpful

and reporting the output.

Thanks,
JDH

Carol -

The ax.set_xlim and ax.set_ylim functions are ignoring the 'equal' status;
not sure why, or whether that is wrong .
But since you are already using pylab, there is a much easier way that does
work.
After your plot(4*cos(an),4*sin(an)) command, simply type

axis('equal')
axis((xmin,xmax,ymin,ymax))  (both -5 and +5 I guess)

and it will work as expected.

If you insist on working the way you were, you have to use the right order,
and fix the limits:

ax.set_xlim(-5, 5)
ax.set_ylim(-5, 5)
ax.set_aspect('equal',fixLimits=3DTrue)

Mark

--__--__--

Hi folks,

I am trying to use quiver with equal axes.  I have not been successful.
 I suspect it is something simple such as not grabbing the correct axes.

If I use figure(figsize(7,7)), the plot seems to be a square, but it is
not.  This is apparent when I run the script using:
 python quiver6.py -dPS
and make two printouts which I then overlay with one rotated by 90 degrees.

I am using matplotlib 0.87.1 with numarray and GTKAgg backend.

Here is my code:

from pylab import *

an =3D linspace(0,2*pi,100)

X,Y =3D meshgrid( arange(-pi,pi,.2),arange(-pi,pi,.2) )
U =3D cos(X)
V =3D sin(Y)

#figure(figsize=3D(7,7))
quiver( X,Y, U, V, color=3DTrue )
plot(4*cos(an),4*sin(an))
ax =3D gca()      # Return the current axes
ax.set_aspect('equal')
ax.set_xlim(-5, 5)
ax.set_ylim(-5, 5)

savefig('quiver6', orientation =3D 'landscape')

--
Ms. Carol A. Leger
SRI International                       Phone: (650) 859-4114
333 Ravenswood Avenue G-273
Menlo Park, CA 94025                    e-mail: le...@sr...

Hi group,

I wonder how I can change the default colors MPL uses? Now, when I
plot 3 arrays in one subplot I get blur, green and red colors on
these. These are nice colors, but I still want to change the defaults.
thx

 >>> matplotlib.__version__
'0.87.2svn'

Python 2.4.1 (#1, Sep 12 2005, 23:33:18)
[GCC 4.0.2 20050901 (prerelease) (SUSE Linux)] on linux2
Type "help", "copyright", "credits" or "license" for more information.
>>> import pylab
Traceback (most recent call last):
  File "<stdin>", line 1, in ?
  File "/usr/lib64/python2.4/site-packages/pylab.py", line 1, in ?
    from matplotlib.pylab import *
  File "/usr/lib64/python2.4/site-packages/matplotlib/pylab.py", line
200, in ?
    from axes import Axes, PolarAxes
  File "/usr/lib64/python2.4/site-packages/matplotlib/axes.py", line 14,
in ?
    from artist import Artist, setp
  File "/usr/lib64/python2.4/site-packages/matplotlib/artist.py", line
4, in ?
    from transforms import identity_transform
  File "/usr/lib64/python2.4/site-packages/matplotlib/transforms.py",
line 193, in ?
    from matplotlib.numerix.linear_algebra import inverse
ImportError: cannot import name inverse


Nils

Hi folks,

I am trying to use quiver with equal axes.  I have not been successful. 
  I suspect it is something simple such as not grabbing the correct axes.

If I use figure(figsize(7,7)), the plot seems to be a square, but it is 
not.  This is apparent when I run the script using:
  python quiver6.py -dPS
and make two printouts which I then overlay with one rotated by 90 degrees.

I am using matplotlib 0.87.1 with numarray and GTKAgg backend.

Here is my code:

from pylab import *

an = linspace(0,2*pi,100)

X,Y = meshgrid( arange(-pi,pi,.2),arange(-pi,pi,.2) )
U = cos(X)
V = sin(Y)

#figure(figsize=(7,7))
quiver( X,Y, U, V, color=True )
plot(4*cos(an),4*sin(an))
ax = gca()	# Return the current axes
ax.set_aspect('equal')
ax.set_xlim(-5, 5)
ax.set_ylim(-5, 5)

savefig('quiver6', orientation = 'landscape')

-- 
Ms. Carol A. Leger
SRI International			Phone: (650) 859-4114
333 Ravenswood Avenue G-273
Menlo Park, CA 94025                    e-mail: le...@sr...