matplotlib-users Mailing List for matplotlib

Hi Phil,

Yes, I can confirm that upgrading fixes the issue. Thanks for the pointer
to cartopy.

Best regards,
Jesper




2014-03-24 12:13 GMT+01:00 Phil Elson <pel...@gm...>:

> I fixed an issue related to this (I too was producing map tiles) in
> matplotlib v1.2 I believe.
>
> The original issue can be found at
> https://github.com/matplotlib/matplotlib/pull/1591 and so I suggest this
> might not be an issue with matplotlib >= v1.3.
>
> Incidentally, if you are producing map tiles you might be interested in
> cartopy which will allow you to produce properly referenced geo maps (and
> therefore tiles) with coastlines etc.
> I've put a short-sh example in a gist () with the rendered results also
> available (https://rawgithub.com/pelson/9738051/raw/map.html). I've also
> got a tornado based handler version which generates the tiles upon HTTP
> request rather than storing the tiles on disk (much more efficient if you
> have highly dynamic data and a caching layer).
>
> Let me know if updating your matplotlib version helps,
>
> Cheers,
>
> Phil
>
>
>
>
>
>
>
> On 24 March 2014 09:45, Jesper Larsen <jes...@gm...> wrote:
>
>> Hi matplotlib users,
>>
>> I am using matplotlib to produce plots (tiles) in a Web Map Service.
>> Unfortunately I cannot get Matplotlib to plot on the entire image. There
>> are one transparent (pixel) line at the bottom and one transparent line at
>> the right. This is of course a problem when the tiles are shown in a map.
>> Please see example below. Can anyone see what I am doing wrong?
>>
>> Best regards,
>> Jesper
>>
>> import numpy as np
>> import matplotlib as mpl
>> from matplotlib.figure import Figure
>> from matplotlib.backends.backend_agg import FigureCanvasAgg as
>> FigureCanvas
>>
>> w = 256
>> h = 256
>> dpi = 128
>> figsize = w/dpi, h/dpi
>> fig = Figure(figsize=figsize, dpi=dpi, frameon=False)
>> canvas = FigureCanvas(fig)
>> ax = fig.add_axes([0, 0, 1, 1])
>>
>> x = np.arange(0, 10, 0.1)
>> y = np.arange(10, 20, 0.2)
>> X, Y = np.meshgrid(x, y)
>> D = np.ones((X.shape[0]-1, X.shape[1]-1))
>> V = np.linspace(0.0, 1.0, 10)
>> ax.pcolor(X, Y, D, antialiased=False)
>> ax.axis( [x[0], x[-1], y[0], y[-1]] )
>> ax.axis('off')
>> filename = 'testfile.png'
>> fig.savefig(filename, dpi=128)
>>
>> # Test image
>> from PIL import Image
>> im = Image.open(filename)
>> print im.getcolors()
>>
>>
>>
>> ------------------------------------------------------------------------------
>> Learn Graph Databases - Download FREE O'Reilly Book
>> "Graph Databases" is the definitive new guide to graph databases and their
>> applications. Written by three acclaimed leaders in the field,
>> this first edition is now available. Download your free book today!
>> http://p.sf.net/sfu/13534_NeoTech
>> _______________________________________________
>> Matplotlib-users mailing list
>> Mat...@li...
>> https://lists.sourceforge.net/lists/listinfo/matplotlib-users
>>
>>
>

Hi Nicolas,

Then everything is transparent. I have no .matplotlibrc file. I pulled the
most recent version of mpl. And that solved the issue.

Best regards,
Jesper



2014-03-24 12:09 GMT+01:00 Nicolas Rougier <Nic...@in...>:

>
>
> If you do not draw at all (no pcolor call), do you still get transparent
> colors ?
> If yes, what is your .matplotlibrc ?
>
>
> Nicolas
>
>
> On 24 Mar 2014, at 11:49, Jesper Larsen <jes...@gm...> wrote:
>
> > Thanks Pierre,
> >
> > from __future__ import division did not help me, I am using mpl 1.1.1rc.
> I will try upgrading to a newer version of mpl and report back whether that
> helps. My output is:
> >
> > [(511, (255, 255, 255, 0)), (65025, (0, 0, 128, 255))]
> >
> > Best regards,
> > Jesper
> >
> >
> >
> > 2014-03-24 11:27 GMT+01:00 Pierre Haessig <pie...@cr...>:
> > Hi,
> >
> > Le 24/03/2014 10:45, Jesper Larsen a écrit :
> > > I am using matplotlib to produce plots (tiles) in a Web Map Service.
> > > Unfortunately I cannot get Matplotlib to plot on the entire image.
> > > There are one transparent (pixel) line at the bottom and one
> > > transparent line at the right. This is of course a problem when the
> > > tiles are shown in a map. Please see example below. Can anyone see
> > > what I am doing wrong?
> > I've run your code and got no transparent pixels.
> >
> > print im.getcolors()
> > [(65536, (0, 0, 128, 255))]
> >
> > I also tried with __future__ division to see if there was something with
> > figsize = w/dpi, h/dpi, but got the same output
> >
> > best,
> > Pierre
> >
> > (python 2.7 on Linux, mpl 1.3.1)
> >
> >
> >
> ------------------------------------------------------------------------------
> > Learn Graph Databases - Download FREE O'Reilly Book
> > "Graph Databases" is the definitive new guide to graph databases and
> their
> > applications. Written by three acclaimed leaders in the field,
> > this first edition is now available. Download your free book today!
> > http://p.sf.net/sfu/13534_NeoTech
> > _______________________________________________
> > Matplotlib-users mailing list
> > Mat...@li...
> > https://lists.sourceforge.net/lists/listinfo/matplotlib-users
> >
> >
> ------------------------------------------------------------------------------
> > Learn Graph Databases - Download FREE O'Reilly Book
> > "Graph Databases" is the definitive new guide to graph databases and
> their
> > applications. Written by three acclaimed leaders in the field,
> > this first edition is now available. Download your free book today!
> >
> http://p.sf.net/sfu/13534_NeoTech_______________________________________________
> > Matplotlib-users mailing list
> > Mat...@li...
> > https://lists.sourceforge.net/lists/listinfo/matplotlib-users
>
>

I fixed an issue related to this (I too was producing map tiles) in
matplotlib v1.2 I believe.

The original issue can be found at
https://github.com/matplotlib/matplotlib/pull/1591 and so I suggest this
might not be an issue with matplotlib >= v1.3.

Incidentally, if you are producing map tiles you might be interested in
cartopy which will allow you to produce properly referenced geo maps (and
therefore tiles) with coastlines etc.
I've put a short-sh example in a gist () with the rendered results also
available (https://rawgithub.com/pelson/9738051/raw/map.html). I've also
got a tornado based handler version which generates the tiles upon HTTP
request rather than storing the tiles on disk (much more efficient if you
have highly dynamic data and a caching layer).

Let me know if updating your matplotlib version helps,

Cheers,

Phil







On 24 March 2014 09:45, Jesper Larsen <jes...@gm...> wrote:

> Hi matplotlib users,
>
> I am using matplotlib to produce plots (tiles) in a Web Map Service.
> Unfortunately I cannot get Matplotlib to plot on the entire image. There
> are one transparent (pixel) line at the bottom and one transparent line at
> the right. This is of course a problem when the tiles are shown in a map.
> Please see example below. Can anyone see what I am doing wrong?
>
> Best regards,
> Jesper
>
> import numpy as np
> import matplotlib as mpl
> from matplotlib.figure import Figure
> from matplotlib.backends.backend_agg import FigureCanvasAgg as FigureCanvas
>
> w = 256
> h = 256
> dpi = 128
> figsize = w/dpi, h/dpi
> fig = Figure(figsize=figsize, dpi=dpi, frameon=False)
> canvas = FigureCanvas(fig)
> ax = fig.add_axes([0, 0, 1, 1])
>
> x = np.arange(0, 10, 0.1)
> y = np.arange(10, 20, 0.2)
> X, Y = np.meshgrid(x, y)
> D = np.ones((X.shape[0]-1, X.shape[1]-1))
> V = np.linspace(0.0, 1.0, 10)
> ax.pcolor(X, Y, D, antialiased=False)
> ax.axis( [x[0], x[-1], y[0], y[-1]] )
> ax.axis('off')
> filename = 'testfile.png'
> fig.savefig(filename, dpi=128)
>
> # Test image
> from PIL import Image
> im = Image.open(filename)
> print im.getcolors()
>
>
>
> ------------------------------------------------------------------------------
> Learn Graph Databases - Download FREE O'Reilly Book
> "Graph Databases" is the definitive new guide to graph databases and their
> applications. Written by three acclaimed leaders in the field,
> this first edition is now available. Download your free book today!
> http://p.sf.net/sfu/13534_NeoTech
> _______________________________________________
> Matplotlib-users mailing list
> Mat...@li...
> https://lists.sourceforge.net/lists/listinfo/matplotlib-users
>
>

If you do not draw at all (no pcolor call), do you still get transparent colors ?
If yes, what is your .matplotlibrc ?


Nicolas


On 24 Mar 2014, at 11:49, Jesper Larsen <jes...@gm...> wrote:

> Thanks Pierre,
> 
> from __future__ import division did not help me, I am using mpl 1.1.1rc. I will try upgrading to a newer version of mpl and report back whether that helps. My output is:
> 
> [(511, (255, 255, 255, 0)), (65025, (0, 0, 128, 255))]
> 
> Best regards,
> Jesper
> 
> 
> 
> 2014-03-24 11:27 GMT+01:00 Pierre Haessig <pie...@cr...>:
> Hi,
> 
> Le 24/03/2014 10:45, Jesper Larsen a écrit :
> > I am using matplotlib to produce plots (tiles) in a Web Map Service.
> > Unfortunately I cannot get Matplotlib to plot on the entire image.
> > There are one transparent (pixel) line at the bottom and one
> > transparent line at the right. This is of course a problem when the
> > tiles are shown in a map. Please see example below. Can anyone see
> > what I am doing wrong?
> I've run your code and got no transparent pixels.
> 
> print im.getcolors()
> [(65536, (0, 0, 128, 255))]
> 
> I also tried with __future__ division to see if there was something with
> figsize = w/dpi, h/dpi, but got the same output
> 
> best,
> Pierre
> 
> (python 2.7 on Linux, mpl 1.3.1)
> 
> 
> ------------------------------------------------------------------------------
> Learn Graph Databases - Download FREE O'Reilly Book
> "Graph Databases" is the definitive new guide to graph databases and their
> applications. Written by three acclaimed leaders in the field,
> this first edition is now available. Download your free book today!
> http://p.sf.net/sfu/13534_NeoTech
> _______________________________________________
> Matplotlib-users mailing list
> Mat...@li...
> https://lists.sourceforge.net/lists/listinfo/matplotlib-users
> 
> ------------------------------------------------------------------------------
> Learn Graph Databases - Download FREE O'Reilly Book
> "Graph Databases" is the definitive new guide to graph databases and their
> applications. Written by three acclaimed leaders in the field,
> this first edition is now available. Download your free book today!
> http://p.sf.net/sfu/13534_NeoTech_______________________________________________
> Matplotlib-users mailing list
> Mat...@li...
> https://lists.sourceforge.net/lists/listinfo/matplotlib-users

Thanks Pierre,

from __future__ import division did not help me, I am using mpl 1.1.1rc. I
will try upgrading to a newer version of mpl and report back whether that
helps. My output is:

[(511, (255, 255, 255, 0)), (65025, (0, 0, 128, 255))]

Best regards,
Jesper



2014-03-24 11:27 GMT+01:00 Pierre Haessig <pie...@cr...>:

> Hi,
>
> Le 24/03/2014 10:45, Jesper Larsen a écrit :
> > I am using matplotlib to produce plots (tiles) in a Web Map Service.
> > Unfortunately I cannot get Matplotlib to plot on the entire image.
> > There are one transparent (pixel) line at the bottom and one
> > transparent line at the right. This is of course a problem when the
> > tiles are shown in a map. Please see example below. Can anyone see
> > what I am doing wrong?
> I've run your code and got no transparent pixels.
>
> print im.getcolors()
> [(65536, (0, 0, 128, 255))]
>
> I also tried with __future__ division to see if there was something with
> figsize = w/dpi, h/dpi, but got the same output
>
> best,
> Pierre
>
> (python 2.7 on Linux, mpl 1.3.1)
>
>
>
> ------------------------------------------------------------------------------
> Learn Graph Databases - Download FREE O'Reilly Book
> "Graph Databases" is the definitive new guide to graph databases and their
> applications. Written by three acclaimed leaders in the field,
> this first edition is now available. Download your free book today!
> http://p.sf.net/sfu/13534_NeoTech
> _______________________________________________
> Matplotlib-users mailing list
> Mat...@li...
> https://lists.sourceforge.net/lists/listinfo/matplotlib-users
>

Hi,

Le 24/03/2014 10:45, Jesper Larsen a écrit :
> I am using matplotlib to produce plots (tiles) in a Web Map Service.
> Unfortunately I cannot get Matplotlib to plot on the entire image.
> There are one transparent (pixel) line at the bottom and one
> transparent line at the right. This is of course a problem when the
> tiles are shown in a map. Please see example below. Can anyone see
> what I am doing wrong?
I've run your code and got no transparent pixels.

print im.getcolors()
[(65536, (0, 0, 128, 255))]

I also tried with __future__ division to see if there was something with
figsize = w/dpi, h/dpi, but got the same output

best,
Pierre

(python 2.7 on Linux, mpl 1.3.1)

Hi matplotlib users,

I am using matplotlib to produce plots (tiles) in a Web Map Service.
Unfortunately I cannot get Matplotlib to plot on the entire image. There
are one transparent (pixel) line at the bottom and one transparent line at
the right. This is of course a problem when the tiles are shown in a map.
Please see example below. Can anyone see what I am doing wrong?

Best regards,
Jesper

import numpy as np
import matplotlib as mpl
from matplotlib.figure import Figure
from matplotlib.backends.backend_agg import FigureCanvasAgg as FigureCanvas

w = 256
h = 256
dpi = 128
figsize = w/dpi, h/dpi
fig = Figure(figsize=figsize, dpi=dpi, frameon=False)
canvas = FigureCanvas(fig)
ax = fig.add_axes([0, 0, 1, 1])

x = np.arange(0, 10, 0.1)
y = np.arange(10, 20, 0.2)
X, Y = np.meshgrid(x, y)
D = np.ones((X.shape[0]-1, X.shape[1]-1))
V = np.linspace(0.0, 1.0, 10)
ax.pcolor(X, Y, D, antialiased=False)
ax.axis( [x[0], x[-1], y[0], y[-1]] )
ax.axis('off')
filename = 'testfile.png'
fig.savefig(filename, dpi=128)

# Test image
from PIL import Image
im = Image.open(filename)
print im.getcolors()