matplotlib-users Mailing List for matplotlib

Is there anyway to set/create legends independent of what I am plotting?
Simply manually create, position and show legends, that not necessarily are
directly linked to the graph.
To be specific, I am plotting several points, in a scatter form,
individually, and depending on the given attributes, I manually set what
colour and marker each point will present. My intention is to create legends
that specify the categories of my data, in a way that I can manually define
what colour/marker they are linked to.
Thanks in advance for any help!
-- 
View this message in context: http://old.nabble.com/Independent-Legends-tp31752112p31752112.html
Sent from the matplotlib - users mailing list archive at Nabble.com.

Hi Jae-Joon,
Thank you for your help. For the time being, I have discovered that using
pdf output and converting to eps using pdf2ps and ps2eps avoids this
problem.
Best,
Bran
Jae-Joon Lee wrote:
> 
> Just in case, I have opened a git issue on this.
> 
> https://github.com/matplotlib/matplotlib/issues/131
> 
> -JJ
> 
> 
> 
> On Wed, Jun 1, 2011 at 1:23 PM, Jae-Joon Lee <lee...@gm...> wrote:
>> I'm not 100% sure on this but it seems that this is a limitation of
>> "psfrag" that the ps backend relies on. The ps backend first produces
>> an eps file without TeX labels, and these TeX labels are put on the
>> eps file with latex+psfrag. And it seems these TeX labels are always
>> above the contents of the eps file.
>>
>> Unfortunately, I don't think this can be easily fixed.
>> Regards,
>>
>> -JJ
>>
>>
>> On Wed, Jun 1, 2011 at 6:34 AM, Brannon <gi...@gm...> wrote:
>>>
>>> Hi,
>>>
>>> I am having trouble with matplotlib 1.0.1 drawing the axis tick labels
>>> over
>>> the legend box in the eps output when useTex is set to true. The plot
>>> shown
>>> after calling plt.show() looks fine, as does the output in pdf, png, svg
>>> etc. Only the postscript appears to be affected.
>>>
>>> The following simple example produces the png and eps files given at the
>>> bottom of this post:
>>>
>>> import matplotlib as mpl
>>> import matplotlib.pyplot as plt
>>> from matplotlib import rc
>>>
>>> rc('text', usetex=True)
>>>
>>> g1 = plt.plot([1,2,3,4],[500,600,700,800])
>>> g2 = plt.plot([1,2,3,4],[700,300,700,200])
>>> g3 = plt.plot([1,2,3,4],[800,600,900,800])
>>>
>>> plt.figlegend([g1,g2,g3],['test1','test2','test3'],'upper left')
>>>
>>> plt.savefig('image.png')
>>> plt.savefig('image.eps')
>>>
>>> plt.show()
>>>
>>> http://old.nabble.com/file/p31744864/image.eps image.eps
>>> http://old.nabble.com/file/p31744864/image.png image.png
>>>
>>> I would very much appreciate some help resolving this issue.
>>>
>>> Thanks
>>> --
>>> View this message in context:
>>> http://old.nabble.com/eps-and-useTex%3A-tick-labels-drawn-over-legend-box-tp31744864p31744864.html
>>> Sent from the matplotlib - users mailing list archive at Nabble.com.
>>>
>>>
>>> ------------------------------------------------------------------------------
>>> Simplify data backup and recovery for your virtual environment with
>>> vRanger.
>>> Installation's a snap, and flexible recovery options mean your data is
>>> safe,
>>> secure and there when you need it. Data protection magic?
>>> Nope - It's vRanger. Get your free trial download today.
>>> http://p.sf.net/sfu/quest-sfdev2dev
>>> _______________________________________________
>>> Matplotlib-users mailing list
>>> Mat...@li...
>>> https://lists.sourceforge.net/lists/listinfo/matplotlib-users
>>>
>>
> 
> ------------------------------------------------------------------------------
> Simplify data backup and recovery for your virtual environment with
> vRanger. 
> Installation's a snap, and flexible recovery options mean your data is
> safe,
> secure and there when you need it. Data protection magic?
> Nope - It's vRanger. Get your free trial download today. 
> http://p.sf.net/sfu/quest-sfdev2dev
> _______________________________________________
> Matplotlib-users mailing list
> Mat...@li...
> https://lists.sourceforge.net/lists/listinfo/matplotlib-users
> 
> 
-- 
View this message in context: http://old.nabble.com/eps-and-useTex%3A-tick-labels-drawn-over-legend-box-tp31744864p31751252.html
Sent from the matplotlib - users mailing list archive at Nabble.com.

I'd really appreciate help on this, even though I'm new to matplotlib and
Nabble.
Consider the following lists:
[1537, 1686, 1858, 2113, 2832]
[1328, 2168]
with max value = 2850
For a colorbar with range 0 to max value and a colour gradient of blue,
green and red, I'd like to show green at y-axis values from the first list
and red at y-axis values from the second list. At any other value the bar
would be blue. Values from one list never occur in the other.
How would I go about accomplishing this?
-- 
View this message in context: http://old.nabble.com/Mapping-colorbar-to-data-tp31748555p31748555.html
Sent from the matplotlib - users mailing list archive at Nabble.com.

PS: One could add to the non-interactive mode part that "pyplot.draw()" has
the same effect as drawing() everything (normally, this does not display
anything, but is necessary so that show() displays the drawn() elements). 
Right?
-- 
View this message in context: http://old.nabble.com/Exact-semantics-of-ion%28%29---tp31728909p31748078.html
Sent from the matplotlib - users mailing list archive at Nabble.com.

Thank you for these precisions.
I think I'm starting to see more clearly what the
interactive/non-interactive modes do with pyplot commands (plot(), draw(),
show(),...), and with draw() methods.
There is only one thing that I'm not sure about: if we look at your script
and leave the ion() were you left it, shouldn't an ax.draw() be called just
before the ion()? (I indeed understand from your last post that one "should
not rely" on "pyplot.figure()" sending a delayed draw() request to the GUI.)
So, to summarize the whole discussion:
* Interactive mode:
- Graph elements plotted with *pyplot* commands (not Matplotlib object
methods) are displayed without the need to use draw() or show().
- However, plots done through Matplotlib objects (like Axes) are normally
not displayed (even though they may be, with some backends). The actual
display of such plots is done through their draw() method (or possibly
pyplot.draw()). This feature might be used for optimization purposes (a
graph can be refreshed on screen once even though multiple updates were
performed on it through Matplotlib object method calls).
- show(), if used, is non-blocking. It displays everything that was drawn()
(for instance figures that were created in non-interactive mode). Things
that were not drawn() might be displayed by some backends, but one should
not rely on this.
* Non-interactive mode
- No figure or graph is displayed and/or refreshed automatically. This is
useful for optimization purposes, when a graph is updated many times before
it reaches its final state (modification of some elements, like changing a
color, etc.).
- show() displays all the drawn() elements and is blocking. It is possible
to switch temporarily to interactive mode in order to make it non-blocking.
=> Is this correct? are there other issues that could be important for
students (and myself!) to grasp so that they use Matplotlib as best as
possible?
-- 
View this message in context: http://old.nabble.com/Exact-semantics-of-ion%28%29---tp31728909p31748057.html
Sent from the matplotlib - users mailing list archive at Nabble.com.

Hello everyone. I am used to plot data with gnuplot, so I can easily
put the figures in a LaTeX document, using the epslatex terminal. For
example:
  file = "data.dat"
  set terminal epslatex
  set output "figure1.tex"
  plot file
http://gnuplot-tricks.blogspot.com/2009/05/gnuplot-tricks-many-say-that-it-is.html#epslatex-terminal
http://www.gnuplotting.org/introduction/output-terminals/#epslatex
This way, two files are generated: one .eps file, which contains the
graphics, and one .tex file, which contains the text. The great
advantage of this is that text is rendered by LaTeX, so the tics,
labels, etc. have the same font as the rest of the document (using the
appropiate packages).
Now I am starting with matplotlib, which has a much nicer API, is more
scriptable and, well, is Python. But, even though I can make
matplotlib render the text with LaTeX, it gets embedded into the image
and I cannot achieve the same advantages I had with gnuplot.
Is there any way I can emulate the epslatex terminal in matplotlib?
Thank you very much!

Just in case, I have opened a git issue on this.
https://github.com/matplotlib/matplotlib/issues/131
-JJ
On Wed, Jun 1, 2011 at 1:23 PM, Jae-Joon Lee <lee...@gm...> wrote:
> I'm not 100% sure on this but it seems that this is a limitation of
> "psfrag" that the ps backend relies on. The ps backend first produces
> an eps file without TeX labels, and these TeX labels are put on the
> eps file with latex+psfrag. And it seems these TeX labels are always
> above the contents of the eps file.
>
> Unfortunately, I don't think this can be easily fixed.
> Regards,
>
> -JJ
>
>
> On Wed, Jun 1, 2011 at 6:34 AM, Brannon <gi...@gm...> wrote:
>>
>> Hi,
>>
>> I am having trouble with matplotlib 1.0.1 drawing the axis tick labels over
>> the legend box in the eps output when useTex is set to true. The plot shown
>> after calling plt.show() looks fine, as does the output in pdf, png, svg
>> etc. Only the postscript appears to be affected.
>>
>> The following simple example produces the png and eps files given at the
>> bottom of this post:
>>
>> import matplotlib as mpl
>> import matplotlib.pyplot as plt
>> from matplotlib import rc
>>
>> rc('text', usetex=True)
>>
>> g1 = plt.plot([1,2,3,4],[500,600,700,800])
>> g2 = plt.plot([1,2,3,4],[700,300,700,200])
>> g3 = plt.plot([1,2,3,4],[800,600,900,800])
>>
>> plt.figlegend([g1,g2,g3],['test1','test2','test3'],'upper left')
>>
>> plt.savefig('image.png')
>> plt.savefig('image.eps')
>>
>> plt.show()
>>
>> http://old.nabble.com/file/p31744864/image.eps image.eps
>> http://old.nabble.com/file/p31744864/image.png image.png
>>
>> I would very much appreciate some help resolving this issue.
>>
>> Thanks
>> --
>> View this message in context: http://old.nabble.com/eps-and-useTex%3A-tick-labels-drawn-over-legend-box-tp31744864p31744864.html
>> Sent from the matplotlib - users mailing list archive at Nabble.com.
>>
>>
>> ------------------------------------------------------------------------------
>> Simplify data backup and recovery for your virtual environment with vRanger.
>> Installation's a snap, and flexible recovery options mean your data is safe,
>> secure and there when you need it. Data protection magic?
>> Nope - It's vRanger. Get your free trial download today.
>> http://p.sf.net/sfu/quest-sfdev2dev
>> _______________________________________________
>> Matplotlib-users mailing list
>> Mat...@li...
>> https://lists.sourceforge.net/lists/listinfo/matplotlib-users
>>
>

I'm not 100% sure on this but it seems that this is a limitation of
"psfrag" that the ps backend relies on. The ps backend first produces
an eps file without TeX labels, and these TeX labels are put on the
eps file with latex+psfrag. And it seems these TeX labels are always
above the contents of the eps file.
Unfortunately, I don't think this can be easily fixed.
Regards,
-JJ
On Wed, Jun 1, 2011 at 6:34 AM, Brannon <gi...@gm...> wrote:
>
> Hi,
>
> I am having trouble with matplotlib 1.0.1 drawing the axis tick labels over
> the legend box in the eps output when useTex is set to true. The plot shown
> after calling plt.show() looks fine, as does the output in pdf, png, svg
> etc. Only the postscript appears to be affected.
>
> The following simple example produces the png and eps files given at the
> bottom of this post:
>
> import matplotlib as mpl
> import matplotlib.pyplot as plt
> from matplotlib import rc
>
> rc('text', usetex=True)
>
> g1 = plt.plot([1,2,3,4],[500,600,700,800])
> g2 = plt.plot([1,2,3,4],[700,300,700,200])
> g3 = plt.plot([1,2,3,4],[800,600,900,800])
>
> plt.figlegend([g1,g2,g3],['test1','test2','test3'],'upper left')
>
> plt.savefig('image.png')
> plt.savefig('image.eps')
>
> plt.show()
>
> http://old.nabble.com/file/p31744864/image.eps image.eps
> http://old.nabble.com/file/p31744864/image.png image.png
>
> I would very much appreciate some help resolving this issue.
>
> Thanks
> --
> View this message in context: http://old.nabble.com/eps-and-useTex%3A-tick-labels-drawn-over-legend-box-tp31744864p31744864.html
> Sent from the matplotlib - users mailing list archive at Nabble.com.
>
>
> ------------------------------------------------------------------------------
> Simplify data backup and recovery for your virtual environment with vRanger.
> Installation's a snap, and flexible recovery options mean your data is safe,
> secure and there when you need it. Data protection magic?
> Nope - It's vRanger. Get your free trial download today.
> http://p.sf.net/sfu/quest-sfdev2dev
> _______________________________________________
> Matplotlib-users mailing list
> Mat...@li...
> https://lists.sourceforge.net/lists/listinfo/matplotlib-users
>

Using AxesGrid does not mean that color scales are shared. It only
takes care of axes placement. And it is your responsibility to sync
colorbars of multiple images.
One option is to use *norm* attribute of images.
 norm = matplotlib.colors.Normalize()
 for i in range(3):
 im = grid[i].imshow(Z, extent=extent, interpolation="nearest",
 norm=norm)
 im = grid[3].imshow(1000 * Z, extent=extent, interpolation="nearest",
 norm=norm)
Regards,
-JJ
On Wed, May 25, 2011 at 11:05 AM, Paul Anton Letnes
<pau...@gm...> wrote:
> Hi!
>
> I am wondering if there is a bug in:
> mpl_toolkits.axes_grid1.AxesGrid
>
> I am trying to run this example (I am of course working on something else, but I am trying to build a minimal example of my problem):
> http://matplotlib.sourceforge.net/examples/axes_grid/demo_axes_grid.html
> This would be a very compact and nice way of showing my results, and I am beginning to love it. There is one thing, though. Given the code attached at the end of this message, the two versions produce figures that look identical. To my mind, however, it doesn't exactly seem that the colorbars are "shared" in any real sense.
>
> Try commenting out the 2-3 lines marked # Version 1 and # Version 2. I attach two plots of what I get: same-looking plots with wildly varying colorbar numbers. I'd like to get the numbers to be the same.
>
> TL;DR: In version 1 I am scaling the data for one of the contour plots by a factor 1000, and would expect that plot to look quite different from the others.
>
> Cheers
> Paul.
>
>
> # CODE:
> import matplotlib
> matplotlib.use('agg')
> import matplotlib.pyplot as plt
> from mpl_toolkits.axes_grid1 import AxesGrid
>
>
> def get_demo_image():
>  import numpy as np
>  from matplotlib.cbook import get_sample_data
>  f = get_sample_data("axes_grid/bivariate_normal.npy", asfileobj=False)
>  z = np.load(f)
>  # z is a numpy array of 15x15
>  return z, (-3,4,-4,3)
>
> def demo_grid_with_single_cbar(fig):
>  """
>  A grid of 2x2 images with a single colobar
>  """
>  grid = AxesGrid(fig, 132, # similar to subplot(132)
>          nrows_ncols = (2, 2),
>          axes_pad = 0.0,
>          share_all=True,
>          label_mode = "L",
>          cbar_location = "top",
>          cbar_mode="single")
>
>  Z, extent = get_demo_image()
>
>  # Version 1
>  #for i in range(3):
>    #im = grid[i].imshow(Z, extent=extent, interpolation="nearest")
>  #im = grid[3].imshow(1000 * Z, extent=extent, interpolation="nearest")
>  # Version 2
>  for i in range(4):
>    im = grid[i].imshow(Z, extent=extent, interpolation="nearest")
>
>  # The rest is the same
>  grid.cbar_axes[0].colorbar(im)
>
>  # This affects all axes as share_all = True.
>  grid.axes_llc.set_xticks([-2, 0, 2])
>  grid.axes_llc.set_yticks([-2, 0, 2])
>
>
> if 1:
>  F = plt.figure(1, (5.5, 2.5))
>
>  F.subplots_adjust(left=0.05, right=0.98)
>
>  demo_grid_with_single_cbar(F)
>
>  plt.savefig('cbars')
>  #plt.draw()
>  #plt.show()
>
>
> ------------------------------------------------------------------------------
> vRanger cuts backup time in half-while increasing security.
> With the market-leading solution for virtual backup and recovery,
> you get blazing-fast, flexible, and affordable data protection.
> Download your free trial now.
> http://p.sf.net/sfu/quest-d2dcopy1
> _______________________________________________
> Matplotlib-users mailing list
> Mat...@li...
> https://lists.sourceforge.net/lists/listinfo/matplotlib-users
>
>