matplotlib-users — Discussion related to using matplotlib

[Matplotlib-users] steps in plot()

[Peter G. <pgr...@ge...>]

Hi:

Want to inquire about gnuplot's 'steps' like functionality in 
matplotlib's plotting. A good example is show here:

http://chem.skku.ac.kr/~wkpark/tutor/gnuplot/gpdocs/steps.htm

In a most hack-ish way possible, for the sake of testing I tried this 
(modified simple_plot.py example):

--------------
#!/usr/bin/env python

from matplotlib.matlab import *

figure(1)
t = arange(0.0, 1.0, 0.01)
s = sin(2*2*pi*t)

t2=concatenate((t[1:], array([t[-1]])), 1)
s2=concatenate((array([s[0]]), s[:-1]), 1)

hlines(s, t, t2)
vlines(t, s, s2)

gca().autoscale_view()   #Using 60.2; need to call this for the the 
parts below y=0 to be visible

xlabel('time (s)')
ylabel('voltage (mV)')
title('About as simple as it gets, folks')'
grid(True)
#axis([0,1,-1,1])
savefig('simple_plot')

show()
-------------

It seems to work quite nicely, but I think an ideal way of doing this 
would be to incorporate it into the plot() command. Perhaps adding an 
option 'steps' (following gnuplot's convention could have steps equal 
'histeps', 'fsteps' or just  'steps' - see link above.. None could mean 
regular plot() behavior).  I would say this would be the most elegant 
option, but probably would call for John (or someone else from the core 
developers) to make the changes. Alternatively we could use above and 
wrap it in a plot_step() function.

Any interest in this? If so which way do we want to go?

Peter

Re: [Matplotlib-users] axes control

[Humufr <hu...@ya...>]

John Hunter wrote:

>>>>>>"Humufr" == Humufr  <hu...@ya...> writes:
>>>>>>            
>>>>>>
>
>    Humufr>           Hello, I would like to know how can I change the
>    Humufr> axes properties for the font of the tick label. I was
>    Humufr> trying to use something like:
>
>    Humufr> xticklabels = get(gca(), 'xticklabels') set(xticklabels,
>    Humufr> 'fontweight', 'bold')
>
>    Humufr> and that's work for some parameter (fontweight for
>    Humufr> example) but it's not possible to choose some others like
>    Humufr> the font type or the angle.
>
>You need to pick the right property names: rotation rather than angle,
>fontname or family for the fontname or family
>
>    from matplotlib.matlab import *
>
>    plot([1,2,3])
>    labels = get(gca(), 'xticklabels')
>    set(labels, fontweight='bold', rotation=45, fontname='cmr10')
>
>    show()
>
>You may want to read http://matplotlib.sourceforge.net/fonts.html and
>the fonts section in the rc file to decide if you want to use the
>family or fontname property.
>
>I think one of the reasons you may be having problem with the fontname
>is that matplotlib appears to fail silently when you specify a font
>that it doesn't find.  In this case, it falls back on a known font.
>We hope to add a verbose feature so you can get some diagnostic
>information about what the font finder is doing.
>

Hi,

Thanks for your answer, I have been able to change the font but the 
things I don't like, no that I think it's not coherent with other 
matplotlib command is that in the font you can't ask fontname=Courier" 
and fontstyle="Italic" (the term angle was bad in my precedent mail :) 
). That doesn't work.
Perhaps it's a problem with my installation (debian and package from 
this distribution and Suse and matplotlib install by hand) but I have an 
error message when I'm trying to have some Italic. I wa trying to use 
the script:

http://matplotlib.sourceforge.net/examples/fonts_demo.py

and I have this message when the script want plot the italic fonts:

Could not match cursive, normal, normal.  Returning 
/usr/share/fonts/truetype/ttf-bitstream-vera/Vera.ttf
Could not match fantasy, normal, normal.  Returning 
/usr/share/fonts/truetype/ttf-bitstream-vera/Vera.ttf
Could not match sans-serif, italic, normal.  Returning 
/usr/share/fonts/truetype/ttf-bitstream-vera/Vera.ttf
Could not match sans-serif, italic, normal.  Returning 
/usr/share/fonts/truetype/ttf-bitstream-vera/Vera.ttf
Could not match sans-serif, italic, normal.  Returning 
/usr/share/fonts/truetype/ttf-bitstream-vera/Vera.ttf

I defined the TTFPATH and I have no change.

I have something else weird with the fonts. I verified i=with Ipython and

from matplotlib.font_manager import fontManager, FontProperties
from matplotlib.matlab import *
p = FontProperties()
p.get_family()

the different fonts available. I obtain:

['Lucida Grande',
 'Verdana',
 'Geneva',
 'Lucida',
 'Bitstream Vera Sans',
 'Arial',
 'Helvetica',
 'sans-serif']

but if I'm trying to use the font Arial and italic in the script that 
give me this message:
 
Could not match sans-serif, italic, normal.  Returning 
/usr/share/fonts/truetype/ttf-bitstream-vera/Vera.ttf

It's seems that the change introduce in the script is not use and that 
matplotlib are using only Vera fonts with no style.




fonts = {
      'color'               : 'k',
      'fontname'            : 'Courier',
      'fontweight'          : 'bold',
      'fontstyle'           : 'italic',
      'fontsize'            : 'xx-large'
      }


ylabel('toto',fonts)

give me exactly the same things than:

fonts = {
      'color'               : 'k',
      'fontname'            : 'Arial,
      'fontweight'          : 'bold',
      'fontstyle'           : 'italic',
      'fontsize'            : 'xx-large'
      }


ylabel('toto',fonts)


The only things that seems to work is the fontweigth.

I have exactly the same things if I'm doing:

t = ylabel('label')
set(t,'fontstyle','italic')


I did probably something wrong but I can't find where so I'm asking for 
some help,

          Nicolas

Re: [Matplotlib-users] show()

[John H. <jdh...@ac...>]

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

    Alan> Is there another way I am supposed to be achieving this
    Alan> behavior?  And can the warning be elaborated a bit, so it is
    Alan> clearer why this is a bad idea?

In truth, the latest release of matplotlib sets a flag on show to
prevent repeated calls from doing any real damage.

But for classroom use, I suggest you just put "interactive : True" in
your rc file and then you have no need for show.  Is there a downside
to this approach?

JDH

[Matplotlib-users] show()

[Alan G I. <ai...@am...>]

We are warned to use show() at most once per script:
 http://matplotlib.sourceforge.net/faq.html#SHOW
For scripts I use in a classroom setting, I generally
need to call show() several times: once for each of
a sequence of figures.  I have been doing this with
the TKAgg backend, and I have had almost no problems.
A figure is shown, and the script does not continue
until I close the figure (by clicking its close button).
This is the behavior I want.

Is there another way I am supposed to be achieving this
behavior?  And can the warning be elaborated a bit, so
it is clearer why this is a bad idea?

Thank you,
Alan Isaac

[Matplotlib-users] linewidth and errorbar

[Humufr <hu...@ya...>]

                      Hello,

I found a problem with the function errorbar. I'm trying to change the 
width of the errorbar. The only way to do this seems to pass by the file 
.matplotlibrc and the default line width (it's not very useful I think 
and an option linewidth will be welcome in the errorbar function). The 
second problem is for the caps who are not change even with the global 
change (see figure in attachement).

       Nicolas

Re: [Matplotlib-users] problem with show()

[John H. <jdh...@ac...>]

>>>>> "Daniel" == Daniel Ferreira <da...@te...> writes:

    Daniel> Hi, I'm learning how to use matplotlib and I'm having a
    Daniel> problem with show().  I did the simplest example of the
    Daniel> Tutorial on the Matplotlib's homepage:

    Daniel> from matplotlib.matlab import * plot([1,2,3,4]) show()

    Daniel> It works fine, but when I try to run it again (second time
    Daniel> on the same session), the plot is not shown. I have
    Daniel> another question too: when I run the show() command, my
    Daniel> program stops and only continues when I close the plot
    Daniel> windows. Is there anyway of show the plot and continues
    Daniel> the program with the plot window openned?

    Daniel> I'm using matplotlib 0.64.4-1 on a Debian GNU/Linux.

Is this the right version number?  I didn't know matplotlib 0.64 was
out yet :-)

    Daniel> Sorry about my poor english.

I didn't even notice - it's not that poor!

The usage of show is a source of confusion to all new users of
matplotlib -- you are not alone.  Please see the FAQ
http://matplotlib.sf.net/faq.html#SHOW.  The main idea is that show is
only meant to be used in matplotlib scripts, not interactively from
the python shell.  If working interactively (type some commands, look
at a figure, type some more commands), you need to set interactive to
be True in your rc file.  Once you have done this (and restarted
python) you won't need to call show at all in an interactive session.
See also http://matplotlib.sf.net/interactive.html and
http://matplotlib.sf.net/.matplotlibrc.

I will try and update the tutorial to make this clearer.

For interactive work, I also recommend you set your backend to TkAgg
in your .matplotlibrc file.  Or use ipython with the pylab option -
google for matplotlib and ipython.

Hope this helps,
JDH

[Matplotlib-users] problem with show()

[Daniel F. <da...@te...>]

Hi,

I'm learning how to use matplotlib and I'm having a problem with show().
I did the simplest example of the Tutorial on the Matplotlib's homepage:

from matplotlib.matlab import *
plot([1,2,3,4])
show()

It works fine, but when I try to run it again (second time on the same
session), the plot is not shown. I have another question too: when I run
the show() command, my program stops and only continues when I close the
plot windows. Is there anyway of show the plot and continues the program
with the plot window openned?

I'm using matplotlib 0.64.4-1 on a Debian GNU/Linux.


Sorry about my poor english.

Thanks.

Daniel.

Re: [Matplotlib-users] xzeroaxis

[Andrew S. <str...@as...>]

John Hunter wrote:

>>>>>>"Alan" == Alan G Isaac <ai...@am...> writes:
>>>>>>            
>>>>>>
>
>    Alan> Now that you have given us axhline as a generalization of
>    Alan> gnuplot's set xzeroaxis please consider letting xticks take
>    Alan> an optional "loc" argument, which could either be "border"
>    Alan> or a number to be interpreted as a 'y' value.
>
>    Alan> So, e.g., axhline() xticks(arange(5),loc=0) would draw an
>    Alan> xzeroaxis and put the xtics along it.
>
>  
>
I tried your suggestion in gnuplot, and I found the results to be ugly 
-- the ticks straddled the axis line, rather than being inside or 
outside.  But I see, in theory, what you mean.

>However, there is a rub.  Sometimes you probably want to place the
>xaxes y location in data units, eg at y=0.  Note that this is not the
>same as y=0 in axes coords, which is always the bottom of the
>rectangle.  This is possible since matplotlib lets you specify the x
>and y data in different data coordinate systems, eg with
>http://matplotlib.sourceforge.net/matplotlib.transforms.html#-blend_xy_sep_transform,
>  
>
Hmm... that URL is not working for me, but I found what you referenced 
at http://matplotlib.sourceforge.net/matplotlib.transforms.html

>but I think it would take some work to make the interface easy and
>intuitive.  The other trick, from the developer side, is to couple the
>location of the ticks and tick labels to the y coord of the axes line.
>Doable, certainly, but would require some redesign because currently
>they all "know" their location.
>  
>
It seems using scipy's traits object for its ability to set observers on 
a variable may be useful here.

As I said earlier, I may take a stab at this, but I always feel as if 
I'm "hacking in the dark" when I try and grasp matplotlib's transform 
and render model...  When/if I look at this in greater detail, I'll 
surely let you know! :)

If you're going to a new render model in the future (kiva?), I wonder 
how useful or long-term would any changes I made now be?

Cheers!
Andrew

Re: [Matplotlib-users] axes control

[John H. <jdh...@ac...>]

>>>>> "Humufr" == Humufr  <hu...@ya...> writes:

    Humufr>           Hello, I would like to know how can I change the
    Humufr> axes properties for the font of the tick label. I was
    Humufr> trying to use something like:

    Humufr> xticklabels = get(gca(), 'xticklabels') set(xticklabels,
    Humufr> 'fontweight', 'bold')

    Humufr> and that's work for some parameter (fontweight for
    Humufr> example) but it's not possible to choose some others like
    Humufr> the font type or the angle.

You need to pick the right property names: rotation rather than angle,
fontname or family for the fontname or family

    from matplotlib.matlab import *

    plot([1,2,3])
    labels = get(gca(), 'xticklabels')
    set(labels, fontweight='bold', rotation=45, fontname='cmr10')

    show()

You may want to read http://matplotlib.sourceforge.net/fonts.html and
the fonts section in the rc file to decide if you want to use the
family or fontname property.

I think one of the reasons you may be having problem with the fontname
is that matplotlib appears to fail silently when you specify a font
that it doesn't find.  In this case, it falls back on a known font.
We hope to add a verbose feature so you can get some diagnostic
information about what the font finder is doing.

    Humufr> Another things I saw in the website, a lot of links are
    Humufr> not working (example:
    Humufr> http://matplotlib.sourceforge.net/figure.html#Axis\ )

That should be
http://matplotlib.sourceforge.net/matplotlib.figure.html#Axis. 

I fixed this (in my local copy).  Let me know if you find some more.

JDH

[Matplotlib-users] axes control

[Humufr <hu...@ya...>]

          Hello,

I would like to know how can I change the axes properties for the font 
of the tick label. I was trying to use something like:

xticklabels = get(gca(), 'xticklabels')
set(xticklabels, 'fontweight', 'bold')

and that's work for some parameter (fontweight for example) but it's not 
possible to choose some others like the font type or the angle.

Another things I saw in the website, a lot of links are not working 
(example: http://matplotlib.sourceforge.net/figure.html#Axis )


Thanks,


       Nicolas

[Matplotlib-users] Pango rotate ?

[Jean-Baptiste C. <jc...@de...>]

S=E6l,


About a year ago John Hunter proposed a routine to rotate pango text ( draw=
_text_rotated)

I wonder if the procedure has been implemented by default in a pango method=
 yet.
It would be VERY useful for many to have the possibility to rotate pangolay=
out, at least by 90,180,270 degrees

Takk

Kv.

Jean-Baptiste


>>>>> "John" =3D=3D John Hunter <jdhunter at ace.bsd.uchicago.edu> writes:

    John> I want to rotate some text 90 degrees.  I'm using pango
    John> layouts and drawable.draw_layout to draw my text.

OK, got this working at the level of manipulating individual pixels in
a gtk.gdk.Image with pixbufs and pixmaps.  For my purposes, making
ylabels, this is probably fast enough, but if anyone has a more
efficient solution, let me know...

However, if the 'Rotate' button is clicked repeatedly, which just
repeats the rotated drawing operation, strange things start to appear
in the region where the rotated text is rendered.  Since repeated
clicks should just be repeating the same drawing operations, it is not
clear to me why this is happening.

Below is an example script.

John Hunter

import pygtk
pygtk.require('2.0')
import gtk
import pango

win =3D gtk.Window()
win.show()
vbox =3D gtk.VBox()
vbox.show()
win.add(vbox)

class C:

    def __init__(self, text):
        self.text =3D text
       =20
    def realize(self, drawingArea):
        self.drawingArea =3D drawingArea
        self.drawable =3D drawingArea.window

        # text properties
        context =3D self.drawingArea.create_pango_context()
        self.layout  =3D self.drawingArea.create_pango_layout(self.text)
        desc =3D pango.FontDescription('Sans 14')
        self.layout.set_font_description(desc)   =20


    def draw_text(self, drawable=3DNone, x=3D100, y=3D100):
        if drawable is None: drawable=3Dself.drawable
       =20
        # draw some text in the foreground color
        gc =3D drawable.new_gc()
        drawable.draw_layout(gc, x=3Dx, y=3Dy, layout=3Dself.layout)
       =20


    def draw_text_rotated(self):
        """
        draw the text to a pixmap, rotate the image, fill a pixbuf
        and draw from the pixbuf
        """

        inkRect, logicalRect =3D self.layout.get_pixel_extents()
        x, y, w, h =3D logicalRect
        winw, winh =3D self.drawable.get_size()
        pixmap =3D gtk.gdk.Pixmap(self.drawable, winw, winh)
        c.draw_text(drawable=3Dpixmap, x=3D0, y=3D0)

        gc =3D pixmap.new_gc()
        if 0:
            # These lines test that the pixmap was drawn to ok
            pixbuf =3D gtk.gdk.Pixbuf(gtk.gdk.COLORSPACE_RGB, 0, 8, w, h)
            pixbuf.get_from_drawable(pixmap, pixmap.get_colormap(),
                                     0, 0, 0, 0, w, h)
            pixbuf.render_to_drawable(self.drawable, gc, 0, 0, 0, 0, w, h, =
0, 0, 0)
            return=20


        imageIn =3D pixmap.get_image(x=3D0, y=3D0, width=3Dw, height=3Dh)
        imageOut =3D gtk.gdk.Image(type=3D0,
                                 visual=3Dpixmap.get_visual(),
                                 width=3Dh, height=3Dw)
        imageOut.set_colormap(imageIn.get_colormap())
        for i in range(w):
            for j in range(h):
                imageOut.put_pixel(j, i, imageIn.get_pixel(w-i-1,j) )


        pixbuf =3D gtk.gdk.Pixbuf(colorspace=3Dgtk.gdk.COLORSPACE_RGB,
                                has_alpha=3D0, bits_per_sample=3D8,
                                width=3Dh, height=3Dw)
        pixbuf.get_from_image(src=3DimageOut, cmap=3DimageIn.get_colormap(),
                              src_x=3D0, src_y=3D0, dest_x=3D0, dest_y=3D0,
                              width=3Dh, height=3Dw)
        pixbuf.render_to_drawable(self.drawable, gc,
                                  src_x=3D0, src_y=3D0,
                                  dest_x=3D50, dest_y=3D100,
                                  width=3Dh, height=3Dw,
                                  dither=3D0, x_dither=3D0, y_dither=3D0)


       =20
       =20
       =20
       =20
c =3D C('Some text')

da =3D gtk.DrawingArea()
da.connect('realize', c.realize)
da.set_size_request(300,300)


da.show()
vbox.pack_start(da, gtk.TRUE, gtk.TRUE)

hbox =3D gtk.HBox()
hbox.show()
vbox.pack_start(hbox, gtk.FALSE, gtk.FALSE)

button =3D gtk.Button('Draw')
button.show()
button.connect('clicked', lambda b: c.draw_text())
hbox.pack_start(button, gtk.TRUE, gtk.TRUE)

button =3D gtk.Button('Rotate')
button.show()
button.connect('clicked', lambda b: c.draw_text_rotated())
hbox.pack_start(button, gtk.TRUE, gtk.TRUE)

button =3D gtk.Button('Quit')
button.show()
button.connect('clicked', lambda b: gtk.mainquit())
hbox.pack_start(button, gtk.TRUE, gtk.TRUE)

gtk.mainloop()
--=20
-----------------------------
Jea...@de...

Department of Statistics
deCODE genetics     Sturlugata,8
570 2993          101 Reykjav=EDk

Re: [Matplotlib-users] problem with x scale on error bar plot

[Gary R. <ga...@em...>]

My comment would be that I agree with Fl=E1vio for the specific case of e=
rrorbar plots. That is, when you have errorbars, you want them slightly i=
nset from the axes so that it's clear where they finish. I think this onl=
y applies to errorbar plots. So, if there's an easy way to have the scali=
ng differentiate, I'd say the endpoints should NOT always exceed the data=
 end points, but should for errorbar plots. I'm think changing the defaul=
t behaviour would upset lots of people.
Another idea might be to just add a command like autoscale(factor) or aut=
oscale(margin) or margin(inset) to specify on a per-case basis.

Gary

----- Original Message -----
From: John Hunter <jdh...@ac...>
Date: Wed, 22 Sep 2004 06:31:58 -0500
To: Fl=E1vio Code=E7o Coelho <fcc...@fi...>
Subject: Re: [Matplotlib-users] problem with x scale on error bar plot

> >>>>> "Fl=E1vio" =3D=3D Fl=E1vio Code=E7o Coelho <fcc...@fi...> =
writes:
>=20
>     Fl=E1vio> Hi john, I am getting a bad autoscale when I generate a
>     Fl=E1vio> two point y-error errorbar plot:
>=20
>     Fl=E1vio> As you can see in the figure attached the points are
>     Fl=E1vio> falling on the margin of the plot.
>=20
>     Fl=E1vio> I think a space before the first point and after the last
>     Fl=E1vio> point of the plot should be added automatically, since no
>     Fl=E1vio> one will want a plot like this by default.
>=20
> This is a fix that is trivial to implement.  But I would like to get
> some feedback because changing the default would affect every almost
> all matplotlib plots.
>=20
> The question is: should the autoscale endpoints always exceed the data
> end points, or are there some cases in which it is desirable for the
> datalim and viewlim to coincide?  In this case
>=20
>     from matplotlib.matlab import *
>     plot([1,2], 'o')
>     show()
>=20
> it seems like you want the viewlim to exceed the datalim.  In cases
> like
>=20
>     from matplotlib.matlab import *
>     t =3D linspace(0,1,100)
>     plot(t, sin(2*pi*t))
>     show()
>=20
> it looks to me like having the xlim at 0,1 is the best solution.
>=20
> To change the behavior, edit matplotlib/ticker.py and search for class
> MultipleLocator.  In the autoscale method, change
>=20
>         vmin =3D self.base.le(dmin)
>         vmax =3D self.base.ge(dmax)
>=20
> to
>         vmin =3D self.base.lt(dmin)
>         vmax =3D self.base.gt(dmax)
>=20
> and try both of the examples above in the le/ge case and in the lt/gt
> case.
>=20
> JDH
>=20
>=20
> -------------------------------------------------------
> This SF.Net email is sponsored by: YOU BE THE JUDGE. Be one of 170
> Project Admins to receive an Apple iPod Mini FREE for your judgement on
> who ports your project to Linux PPC the best. Sponsored by IBM.
> Deadline: Sept. 24. Go here: http://sf.net/ppc_contest.php
> _______________________________________________
> Matplotlib-users mailing list
> Mat...@li...
> https://lists.sourceforge.net/lists/listinfo/matplotlib-users

--=20
___________________________________________________________
Sign-up for Ads Free at Mail.com
http://promo.mail.com/adsfreejump.htm

Re: [Matplotlib-users] Printing from WX

[Chris B. <Chr...@no...>]

John Hunter wrote:

> Printing is typically platform dependent

wxPython provides a platform independent printing framework. It 
certainly should be possible to make use of it in the wx Back end. When 
I have a need for it, I may take a stab at it.

-Chris



-- 
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...

Re: [Matplotlib-users] xzeroaxis

[John H. <jdh...@ac...>]

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

    Alan> Now that you have given us axhline as a generalization of
    Alan> gnuplot's set xzeroaxis please consider letting xticks take
    Alan> an optional "loc" argument, which could either be "border"
    Alan> or a number to be interpreted as a 'y' value.

    Alan> So, e.g., axhline() xticks(arange(5),loc=0) would draw an
    Alan> xzeroaxis and put the xtics along it.

    Alan> Am I overlooking an obvious way to do this (in the matlab
    Alan> module)?


Andrew Straw contacted me earlier about a related issue, that is, to
be able to offset the x axis line, ticks, and label away from the data
region, as in
http://dickinson.caltech.edu/research_visualflightcontrol.html.

I think this is a generally useful thing.  Currently, the axes border
is drawn as a rectangle, with a facecolor and the border determined by
the rectangle edgecolor.  The ticks are then placed on the rectangle
border.

The proposal is to change this behavior, so that the rectangle border
edgecolor becomes the same as the facecolor (ie invisible border).
The border is then replaced by four lines, call them bottom, top, left
and right.  Each of these four lines could be placed independently in
axes coords: 0,0,1,1 = left, bottom, right, top.  Eg, the bottom line
would default to

  bottom = Line2D([0,1], [0,0], transform=ax.transAxes)

ie it the x range would stretch from 0-1 (left-right) and the y coord
would be 0 (bottom) in axes coords.

To offset this line below the data range (the white rectangle defined
by the axes box) you could, for example, make the ycoord -0.1.  To
place the x axes in the middle of the axes box, you would make the
ycoord 0.5.

However, there is a rub.  Sometimes you probably want to place the
xaxes y location in data units, eg at y=0.  Note that this is not the
same as y=0 in axes coords, which is always the bottom of the
rectangle.  This is possible since matplotlib lets you specify the x
and y data in different data coordinate systems, eg with
http://matplotlib.sourceforge.net/matplotlib.transforms.html#-blend_xy_sep_transform,
but I think it would take some work to make the interface easy and
intuitive.  The other trick, from the developer side, is to couple the
location of the ticks and tick labels to the y coord of the axes line.
Doable, certainly, but would require some redesign because currently
they all "know" their location.

JDH

Re: [Matplotlib-users] problem with x scale on error bar plot

[John H. <jdh...@ac...>]

>>>>> "Fl=E1vio" =3D=3D Fl=E1vio Code=E7o Coelho <fcc...@fi...> wr=
ites:

    Fl=E1vio> Hi john, I am getting a bad autoscale when I generate a
    Fl=E1vio> two point y-error errorbar plot:

    Fl=E1vio> As you can see in the figure attached the points are
    Fl=E1vio> falling on the margin of the plot.

    Fl=E1vio> I think a space before the first point and after the last
    Fl=E1vio> point of the plot should be added automatically, since no
    Fl=E1vio> one will want a plot like this by default.

This is a fix that is trivial to implement.  But I would like to get
some feedback because changing the default would affect every almost
all matplotlib plots.

The question is: should the autoscale endpoints always exceed the data
end points, or are there some cases in which it is desirable for the
datalim and viewlim to coincide?  In this case

    from matplotlib.matlab import *
    plot([1,2], 'o')
    show()

it seems like you want the viewlim to exceed the datalim.  In cases
like

    from matplotlib.matlab import *
    t =3D linspace(0,1,100)
    plot(t, sin(2*pi*t))
    show()

it looks to me like having the xlim at 0,1 is the best solution.

To change the behavior, edit matplotlib/ticker.py and search for class
MultipleLocator.  In the autoscale method, change

        vmin =3D self.base.le(dmin)
        vmax =3D self.base.ge(dmax)

to
        vmin =3D self.base.lt(dmin)
        vmax =3D self.base.gt(dmax)

and try both of the examples above in the le/ge case and in the lt/gt
case.

JDH

Re: [Matplotlib-users] performance?

[John H. <jdh...@ac...>]

>>>>> "Wayne" == Wayne Christopher <wa...@4r...> writes:

    Wayne> I have a general performance question - is matplotlib
    Wayne> expected to deal well with large plots when you are zoomed
    Wayne> into a small region?  That is, if I have 100K points in a
    Wayne> graph but have set my limits so that only 100 of them are
    Wayne> visible at one time, should that be comparable to creating
    Wayne> a plot with only the visible items and re-creating it each
    Wayne> time the limits are moved?  (Assuming the X axis is
    Wayne> monotonic and I do the obvious type of filtering in
    Wayne> Python.)  It seems to be a bit sluggish...

There is a data filtering mode already built in, which is designed for
the case, as you suggest, that x is monotonic.  To use it, you set the
property 'data_clipping=True' as in 

  plot(x, y, data_clipping=True)

However, I tried it in your test case and it doesn't help a lot.  The
reason is the numeric clip is too time consuming.  However, I think I
know a good fix.  In the current implementation, the data is clipped
to the view limits, and every time the view limits are moved the clip
has to be performed again.  It would be better to clip to for example

    vmax, vmin = ax.get_xlim()
    interval = vmax-vmin
    clipmin = vmin-interval
    clipmax = vmax+interval

and only update the clip lim when a new vlim / vmax exceeds them.
Then you would only have top perform the clip occasionally, which
would be a big when if you have a large x range and have zoomed into a
narrow part of the interval.

I'll take a stab at it.

JDH

Re: [Matplotlib-users] ps / eps

[John H. <jdh...@ac...>]

>>>>> "Humufr" == Humufr  <hu...@ya...> writes:

    Humufr>                 Hello, I would like to know if it's
    Humufr> possible to have only the figure save in postscript. An
    Humufr> eps file in some sort with not white above or at the
    Humufr> bottom of the page?

Yes, you simply need 

  savefig('somefile.eps') 

at the end of your script.

If you are using a GUI dialog box, you just specify the eps extension
on the filename.

JDH

Re: [Matplotlib-users] Printing from WX

[John H. <jdh...@ac...>]

>>>>> "Gary" ==   <ga...@op...> writes:

    Gary> Hi all, I am trying to incorporate viewing of charts and the
    Gary> printing of the viewed chart from an application written in
    Gary> WxPython V2.5 and Python V2.3 My preference at the momment
    Gary> is to use the EMBEDDED method of displaying the charts in my
    Gary> Python app.  I want to have a print button that the user can
    Gary> just click to get a printout of the chart.

    Gary> Can anyone point me to somewhere that tells me how to print
    Gary> a displayed chart.

Printing is typically platform dependent, which you don't mention.
Here is a solution for embedded wxagg on win32

  http://sourceforge.net/mailarchive/message.php?msg_id=9061267

Let me know if this works for you...

JDH

[Matplotlib-users] Printing from WX

[<ga...@op...>]

Hi all,

I am trying to incorporate viewing of charts and the printing of the =
viewed chart from an application written in WxPython V2.5 and Python =
V2.3
My preference at the momment is to use the EMBEDDED method of displaying =
the charts in my Python app.  I want to have a print button that the =
user can just click to get a printout of the chart.

Can anyone point me to somewhere that tells me how to print a displayed =
chart.

Thanks
Gary Giles

[Matplotlib-users] performance?

[Wayne C. <wa...@4r...>]

I have a general performance question - is matplotlib expected to deal well with 
  large plots when you are zoomed into a small region?  That is, if I have 100K 
points in a graph but have set my limits so that only 100 of them are visible at 
one time, should that be comparable to creating a plot with only the visible 
items and re-creating it each time the limits are moved?  (Assuming the X axis 
is monotonic and I do the obvious type of filtering in Python.)  It seems to be 
a bit sluggish...

Thanks,

	Wayne

Re[2]: [Matplotlib-users] color

[Alan G I. <ai...@am...>]

> Alan G Isaac wrote:
>> It would be great to be able to, e.g.,
>> matlab.plot(d,f(d),'(1,1,0)-')
>> or
>> matlab.plot(d,f(d),'#eeefff-')

On Tue, 21 Sep 2004, Gary apparently wrote:
> I second the request.
> But I have an idea.  Try this:
> matlab.plot(d, f(d), color = (1,0,0))


In retrospect it is obvious.
Better yet, you can
matlab.plot(d, f(d), '--', color = (1,0,0))
etc.

Excellent!

Thank you!
Alan Isaac

[Matplotlib-users] ps / eps

[Humufr <hu...@ya...>]

                Hello,

I would like to know if it's possible to have only the figure save in 
postscript. An eps file in some sort with not white above or at the 
bottom of the page?

Thanks,

    Nicolas

Re[2]: [Matplotlib-users] color

[Alan G I. <ai...@am...>]

On Tue, 21 Sep 2004, Gary apparently wrote:
> import matplotlib.colors
> help(matplotlib.colors)

Yes, I've been using this to set arbitrary
colors everywhere except in the plot command.
Maybe I'm missing something obvious, but the
3rd argument to a plot command must be a string,
and despite the apparent suggest in the docs,
hex strings are not supported. 

That is, I cannot find a way to do this.

It would be great to be able to, e.g.,
matlab.plot(d,f(d),'(1,1,0)-')
or
matlab.plot(d,f(d),'#eeefff-')

Am I missing an obvious way to do this?

fwiw,
Alan

Re: [Matplotlib-users] turn off antialiasing

[Flavio C. C. <fcc...@ci...>]

A quick google search reveals that someone has already published a
tutorial for animating graphs made with matplotlib using mencoder...
Check it out:
http://web.media.mit.edu/~lifton/code/GraphMovies/

He even provides an example script...

Synchronicity...

Cheers,

Flavio



On Mon, 2004-09-20 at 20:50, John Hunter wrote:

> >>>>> "Peter" == Peter Groszkowski <pgr...@ge...> writes:
> 
>     Peter> use mplayer to put the png images together.. results are
>     Peter> quite nice...
> 
> Can you post the command you use to play the pngs?  Can mplayer output
> an avi or mpg, or does it just animate the images.
> 
> I saw from the man pages that something like
> 
>  mplayer mf=type=png:fps=25 mf:/filemask
> 
> is required, but I don't know how to set the filemask of if this is
> the right approach.  I'd like to add some instructions to the FAQ.
> 
> Thanks,
> JDH
> 
> 
> -------------------------------------------------------
> This SF.Net email is sponsored by: YOU BE THE JUDGE. Be one of 170
> Project Admins to receive an Apple iPod Mini FREE for your judgement on
> who ports your project to Linux PPC the best. Sponsored by IBM.
> Deadline: Sept. 24. Go here: http://sf.net/ppc_contest.php
> _______________________________________________
> Matplotlib-users mailing list
> Mat...@li...
> https://lists.sourceforge.net/lists/listinfo/matplotlib-users

[Matplotlib-users] controlling borders (axes)

[Alan G I. <ai...@am...>]

matplotlib.matlab allows control of axis (border) display with
axis('off')
Is finer control offered?  
If not, please consider the following an enhancement
request.

I like the behavior of gnuplot,
where borders are encoded in a 12-bit integer.
The bottom four bits control the border for a 2D plot.
            Bit   include plot axis/border:
              1   bottom 
              2   left  
              4   top  
              8   right

I find it possible to mimic this using the new
axhline and axvline to a certain extent (but, 
e.g., no tics). However a axvline placed at
ymin tends not to display on screen unless linewidth is
increased.

fwiw,
Alan Isaac