matplotlib-users Mailing List for matplotlib

Hi,

I have a problem with subplot_adjust but I don't know if it's a bug or because 
I didn't understand how to use it...

I would like to create a 2 by 3 plots without space between the subplot and I 
can't arrive to manage it...


Is the results of the next script is normal?

 import pylab

fig1 = pylab.figure()
fig1.subplots_adjust(wspace=0,hspace=0)

pylab.subplot(231)
pylab.subplot(232)
pylab.subplot(233)
pylab.subplot(234)
pylab.subplot(235)
pylab.subplot(236)

fig2 = pylab.figure()
fig2.subplots_adjust(wspace=0)

pylab.subplot(231)
pylab.subplot(232)
pylab.subplot(233)
pylab.subplot(234)
pylab.subplot(235)
pylab.subplot(236)

fig3 = pylab.figure()
fig3.subplots_adjust(hspace=0)

pylab.subplot(231)
pylab.subplot(232)
pylab.subplot(233)
pylab.subplot(234)
pylab.subplot(235)
pylab.subplot(236)

pylab.show()

Thanks,

N.

Once the axes are the same, can one get the actual bars to align? hist 
() arranges them to look well in their original ranges, so they don't  
line up together, AFAICT:

#plotting barcharts w/different ranges on same axis
import pylab
a = [1]*2 + [2]*3 + [3]*4
b = [3]*1 + [4]*2 + [5]*3
allim = (min(min(a),min(b)), max(max(a),max(b)))

top = pylab.subplot(211)
vala, bina, patcha = pylab.hist(a)
#top.set_xlim(allim) #doesn't work here
bot = pylab.subplot(212, sharex=top)
valb, binb, patchb = pylab.hist(b)
top.set_xlim(allim)
pylab.show()

I'm using the class API with TkAgg (and matplotlib 0.87.7) to plot 
cumulative data with a known X range and an unknown Y range. The basic 
code is appended. In summary: it all happens on one Axis on one Figure 
and most of it happens with one Line. I'm running into a few problems:

- If I use frameon=False when creating my Figure then when I set the 
plot X scale the old scale is also shown, resulting in a messy display. 
Two sets of tick labels (old and new) overlap. Simply using frameon=True 
prevents this. Is this a bug or a feature?

- I initially tried frameon=False because the plot frame contains a 
1/16" white border that cuts off the edges of the text of the axis 
labels. Is there an easy way to get rid of that white border?

- I plot data points as they slowly arrive. To do this I get a Line:
        plotLine = self.plotAxis.plot([], [], 'bo')[0]
and then as data comes in, I use set_data to update it:
  plotLine.set_data(focPosArr[:measInd], fwhmArr[:measInd])
This works except that autoscale does *not* occur so I have to manually 
set the Y scale each time.

I am guessing autoscaling is only performed as part of a plot command, 
not as a result of set_data, and perhaps this is a feature (potentially 
useful if lots of data comes in quickly). If so, is there a workaround 
that is preferable to manually setting the Y scale each time?

-- Russell

P.S. I set things up this way:
        plotFig = Figure(figsize=(4,1), frameon=True)
        self.figCanvas = FigureCanvasTkAgg(plotFig, sr.master)
        col = gr.getNextCol()
        row = gr.getNextRow()
        self.figCanvas.get_tk_widget().grid(row=0, column=col, 
rowspan=row, sticky="news")
        self.plotAxis = plotFig.add_subplot(1, 1, 1)

Then for every "run" of this code I initialize the plot this way:
        self.plotAxis.clear()
        self.plotAxis.set_xlabel("Focus Offset (microns)")
        self.plotAxis.set_ylabel("FWHM (pixels)")
        self.plotAxis.grid(True)
        #self.plotAxis.set_autoscale_on(True)
        self.plotAxis.autoscale_view(scalex=False, scaley=True)
        plotLine = self.plotAxis.plot([], [], 'bo')[0]
        self.plotAxis.set_xlim((minFoc, maxFoc))
        self.figCanvas.draw()

(In not sure if both autoscale commands are needed; the manual is pretty 
vague on this point).

On Wednesday, November 01, 2006, at 02:17PM, "John Hunter" <jdh...@ac...> wrote:
>>>>>> "listservs" == listservs  <lis...@ma...> writes:
>
>    listservs> I am trying to produce a series of histograms of
>    listservs> related data, for which I want the ranges and scales of
>    listservs> the x-axes to be the same.  However, I dont see an
>    listservs> obvious way of doing this with hist, since specifying
>    listservs> nbins will not guarantee the same axis for each. Is
>    listservs> there some sort of hack that I can use to do this (or
>    listservs> an argument I am overlooking?).
>
>
>You can force axes in the same figure to share the same xaxes
>
>ax1 = subplot(211)
>ax2 = subplot(212, sharex=ax1)
>
>then any changes (panning and zooming) in one will be instantaneously
>reflected in the other.
>
>For axes in different figures, you can set the xlim manually
>
>ax1.set_xlim(xmin, xmax)
>ax2.set_xlim(xmin, xmax)
>

Thanks John,

I think I almost have it, but not quite. Say the first dataset is in the range 2-4, and the second is in the range 1-3. At the moment, I am doing as you say, and what happens is that the x axis gets defined by the first plot, and given to the second with the sharex argument. However, this ends up giving the second plot a range of 2-4, so much of the data (from 1-2) is not shown. What I need is for the axes of _both_ plots to be expanded to the range 1-4. 

C.

>>>>> "listservs" == listservs  <lis...@ma...> writes:

    listservs> I am trying to produce a series of histograms of
    listservs> related data, for which I want the ranges and scales of
    listservs> the x-axes to be the same.  However, I dont see an
    listservs> obvious way of doing this with hist, since specifying
    listservs> nbins will not guarantee the same axis for each. Is
    listservs> there some sort of hack that I can use to do this (or
    listservs> an argument I am overlooking?).


You can force axes in the same figure to share the same xaxes

ax1 = subplot(211)
ax2 = subplot(212, sharex=ax1)

then any changes (panning and zooming) in one will be instantaneously
reflected in the other.

For axes in different figures, you can set the xlim manually

ax1.set_xlim(xmin, xmax)
ax2.set_xlim(xmin, xmax)

There is more sophisticated stuff you can do with observers and event
handling to couple axes between figures in the presence of panning and
zooming if need be.

JDH

>>>>> "Asrarahmed" == Asrarahmed Kadri <aj...@go...> writes:

    Asrarahmed> Hi folks, I am interested in using Matplotlib for
    Asrarahmed> drawing bar charts.  Is it possible to load data from
    Asrarahmed> a file and use it for drawing the barchart.

Yes -- see the following examples.  The first shows you how to use
"load" to load data from a file

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

and this shows you how to make a barchart

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

Merging these two techniques is an exercise left to the reader...

JDH

I am trying to produce a series of histograms of related data, for  
which I want the ranges and scales of the x-axes to be the same.  
However, I dont see an obvious way of doing this with hist, since  
specifying nbins will not guarantee the same axis for each. Is there  
some sort of hack that I can use to do this (or an argument I am  
overlooking?).

Thanks,
--
Christopher Fonnesbeck
fon...@ma...

Hi folks,

I am interested in using Matplotlib for drawing bar charts.
Is it possible to load data from a file and use it for drawing the barchart.

Thanks in anticipation.


Regards,
Asrarahmed


-- 
To HIM you shall return.

Le Mardi 31 Octobre 2006 23:47, Robert Kern a =E9crit=A0:
> hu...@ya... wrote:
> > I tried:
> >
> > label=3D'$\textrm{test}_2$
> > xlabel(r label)
> > xlabel(r+label)
> >
> > etc but it not working (like I expected). So I would like to know if
> > there are a way to precise that the text is a raw string by another thi=
ng
> > that the r character just before the string. Perhaps that will be good =
to
> > have an option like raw=3Dtrue or something similar?
>
> There is no such thing as a raw string *object*. There are only raw string
> *literals*. The r'' determines how the source code is parsed, not how the
> contents of the object is treated.
>
>    label =3D r'$\textrm{test}_2$'
>    xlabel(label)
>
> After the source code containing the string literal is parsed, the string
> is simply a string.

Thank you very much I didn't know this. That will be very helpful.

N.

hu...@ya... wrote:
> I tried:
> 
> label='$\textrm{test}_2$
> xlabel(r label)
> xlabel(r+label)
> 
> etc but it not working (like I expected). So I would like to know if there are 
> a way to precise that the text is a raw string by another thing that the r 
> character just before the string. Perhaps that will be good to have an option 
> like raw=true or something similar?

There is no such thing as a raw string *object*. There are only raw string 
*literals*. The r'' determines how the source code is parsed, not how the 
contents of the object is treated.

   label = r'$\textrm{test}_2$'
   xlabel(label)

After the source code containing the string literal is parsed, the string is 
simply a string.

-- 
Robert Kern

"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

Le Mardi 31 Octobre 2006 23:14, Eric Firing a =E9crit=A0:
> hu...@ya... wrote:
> > Hi,
> >
> > I have different questions to use tex with matplotlib.
> >
> > The first one:
> >
> > If I'm doing:
> >
> > rc('text', usetex=3DTrue)
> > xlabel('$\textrm{toto}$')  # bad
> > xlabel(r'$\textrm{toto}$') # ok
> >
> > The results are not the same and I don't understand why.
> >
> > and for:
> >
> > xlabel('$\textit{toto}$')  #bad
> > xlabel(r'$\textit{toto}$') #ok
> >
> > same things, but it's working fine for:
> >
> > xlabel('$\it{toto}$') #ok
> > xlabel(r'$\it{toto}$') #ok
> >
> > but it's not working for \bf:
> >
> > xlabel('$\textbf{toto}$')  # bad
> > xlabel(r'$\textbf{toto}$') # ok
> > xlabel('$\bf{toto}$')        # bad
> > xlabel(r'$\bf{toto}$')       # ok
> >
> > And this bring to the second question, I would like to create the xlabel
> > with latex stuf inside somewhere in my script:
> >
> > label=3D'$\textrm{test}_2$'
> >
> > but that can't work:
> >
> > xlabel(label) # bad
> >
> > How to tell that I want the "r" before the chains to have TeX working
> > like expected?
>
> I think you have much to gain and nothing to lose by *always* using raw
> strings for TeX.  The reason your "bad" cases are bad is that \b and \t
> are escape sequences; in an ordinary string, without the leading r, the
> two-character sequences are translated into control characters (bell and
> tab, respectively).  \i is not an escape sequence, so it gets left alone.
>
> Eric

I agree with you but how can I use the raw strings when I want to compose a=
=20
text (2nd question)

I tried:

label=3D'$\textrm{test}_2$
xlabel(r label)
xlabel(r+label)

etc but it not working (like I expected). So I would like to know if there =
are=20
a way to precise that the text is a raw string by another thing that the r=
=20
character just before the string. Perhaps that will be good to have an opti=
on=20
like raw=3Dtrue or something similar?

Thanks,

N.

hu...@ya... wrote:
> Hi,
> 
> I have different questions to use tex with matplotlib.
> 
> The first one:
> 
> If I'm doing:
> 
> rc('text', usetex=True)
> xlabel('$\textrm{toto}$')  # bad
> xlabel(r'$\textrm{toto}$') # ok
> 
> The results are not the same and I don't understand why.
> 
> and for:
> 
> xlabel('$\textit{toto}$')  #bad
> xlabel(r'$\textit{toto}$') #ok
> 
> same things, but it's working fine for:
> 
> xlabel('$\it{toto}$') #ok
> xlabel(r'$\it{toto}$') #ok
> 
> but it's not working for \bf:
> 
> xlabel('$\textbf{toto}$')  # bad
> xlabel(r'$\textbf{toto}$') # ok
> xlabel('$\bf{toto}$')        # bad
> xlabel(r'$\bf{toto}$')       # ok
> 
> And this bring to the second question, I would like to create the xlabel with 
> latex stuf inside somewhere in my script:
> 
> label='$\textrm{test}_2$'
> 
> but that can't work:
> 
> xlabel(label) # bad
> 
> How to tell that I want the "r" before the chains to have TeX working like 
> expected?

I think you have much to gain and nothing to lose by *always* using raw 
strings for TeX.  The reason your "bad" cases are bad is that \b and \t 
are escape sequences; in an ordinary string, without the leading r, the 
two-character sequences are translated into control characters (bell and 
tab, respectively).  \i is not an escape sequence, so it gets left alone.

Eric

Hi there.  I have the following versions of numpy and matplotlib
installed on Debian sarge:

In [3]: matplotlib.__version__
Out[3]: '0.87.7'

In [4]: numpy.__version__
Out[4]: '1.0.1.dev3422'

Recent installations from SVN started giving me a segmentation fault
error when I use pylab.close().  For example, I want to press 'q' to
close the plot window interactively, but I get

$ ./test.py
Segmentation fault

when I do that (test.py is attached below).  When I click on the close
button on the window widget (i.e., Alt+F4), the segfault does NOT
happen.  Is this a known bug?  I'd appreciate your help!

Taro

--BEGIN test.py-----------
#!/usr/bin/env python
import pylab as P

def on_press(event):
    key = event.key
    if key == 'q':
        P.close()

def main():
    P.plot([1,2,3])
    P.connect('key_press_event',on_press)
    P.show()
    print 'done.'

if __name__ == '__main__':
    main()

--END test.py---------