That looks fine to me too, but if you plot that as one subplot in a 5x5 array
of subplots or more, then you can see the shift I am talking about in the
eps file.  Example:

import matplotlib.pyplot as plt
import numpy as np

arr = np.zeros((11, 11), dtype="d")
arr[3,3]=1
plt.figure(1)
plt.subplot(10,10,1)
im = plt.imshow(arr, interpolation="nearest", origin="lower")
cont = plt.contour(arr, levels=[0.5])
plt.savefig("a.eps")


Jae-Joon Lee wrote:
> 
> I tried a simple array (see the code below) but cannot reproduce the
> problem you reported.
> 
> import matplotlib.pyplot as plt
> import numpy as np
> 
> arr = np.zeros((11, 11), dtype="d")
> arr[3,3]=1
> im = plt.imshow(arr, interpolation="nearest", origin="lower")
> cont = plt.contour(arr, levels=[0.5])
> plt.savefig("a.eps")
> 
> Do you still see the shift with the above example code?
> And what version of matplotlib are you using?
> 
> If possible, please post a complete script with the data (use a
> mock-up data if you want)?
> Regards,
> 
> -JJ
> 
> 
> 
> On Fri, Jul 23, 2010 at 12:16 AM, Jenna Lemonias
> <je...@astro.columbia.edu> wrote:
>> No, I don't think the issue is a flip in the y-axis.  I have a number of
>> different examples of this, and many in which the contour is an ellipse
>> so I
>> can tell that the overall positioning is correct.  It seems like
>> something
>> is going wrong only when I save the image...  Thanks for the suggestion
>> though!
>>
>> Jenna
>>
>> On Wed, Jul 21, 2010 at 3:40 PM, Jenna Lemonias
>> <je...@astro.columbia.edu>wrote:
>>
>>> I am trying to save a matplotlib 2d array image with an overlaid contour
>>> as
>>> an eps file.  The contour appears to be shifted with respect to the
>>> image
>>> underneath in the eps file, particularly when I zoom in on the image.
>>> This
>>> shift is not noticeable in the plot within matplotlib.
>>>
>>> I am using imshow to display the image.  The contour is created by
>>> plotting
>>> a list of closely-spaced x,y coordinates.  The attached file
>>> matplotlib.png
>>> is a screenshot of the (zoomed-in) image as displayed by matplotlib. 
>>> The
>>> attached file epsfile.png is a screenshot of the (zoomed-in) eps file.
>>> When
>>> I save this image as an eps file, it is actually 1 of 20 subplots and
>>> the
>>> shift is noticeable in each subplot.
>>>
>>> Thanks in advance for your help!
>>>
>>> Jenna
>>>
>>>
>> Just as a wild guess, could this actually be an issue with how imshow
>> uses
>> the upper-left corner for (0,0)?  I have seen 1-pixel shifts before, but
>> this shift is a little dramatic and I am left wondering if what we are
>> really seeing is that the contour that is desired should actually be
>> fliped
>> in the y-axis?
>>
>> Maybe you could try another example where you try to draw a contour
>> further
>> away from the center of the image and see if it still goes in the spot
>> you
>> expect it to be?
>>
>> Ben Root
>>
>> ------------------------------------------------------------------------------
>> This SF.net email is sponsored by Sprint
>> What will you do first with EVO, the first 4G phone?
>> Visit sprint.com/first -- http://p.sf.net/sfu/sprint-com-first
>> _______________________________________________
>> Matplotlib-users mailing list
>> Matplotlib-users@lists.sourceforge.net
>> https://lists.sourceforge.net/lists/listinfo/matplotlib-users
>>
>>
> 
> ------------------------------------------------------------------------------
> The Palm PDK Hot Apps Program offers developers who use the
> Plug-In Development Kit to bring their C/C++ apps to Palm for a share 
> of $1 Million in cash or HP Products. Visit us here for more details:
> http://ad.doubleclick.net/clk;226879339;13503038;l?
> http://clk.atdmt.com/CRS/go/247765532/direct/01/
> _______________________________________________
> Matplotlib-users mailing list
> Matplotlib-users@lists.sourceforge.net
> https://lists.sourceforge.net/lists/listinfo/matplotlib-users
> 
> 

-- 
View this message in context: 
http://old.nabble.com/Saving-as-eps-file-shifts-image--tp29232680p29288310.html
Sent from the matplotlib - users mailing list archive at Nabble.com.


------------------------------------------------------------------------------
The Palm PDK Hot Apps Program offers developers who use the
Plug-In Development Kit to bring their C/C++ apps to Palm for a share
of $1 Million in cash or HP Products. Visit us here for more details:
http://p.sf.net/sfu/dev2dev-palm
_______________________________________________
Matplotlib-users mailing list
Matplotlib-users@lists.sourceforge.net
https://lists.sourceforge.net/lists/listinfo/matplotlib-users

Reply via email to