[Numpy-discussion] scipy image processing memory leak in python 2.7?

Joseph McGlinchy JMcGlinchy at esri.com
Wed Jan 29 13:19:16 EST 2014


I have updated to scipy 0.13.2 (specifically scipy-0.13.2-win32-superpack-python2.7.exe (59.1 MB)<http://sourceforge.net/projects/scipy/files/latest/download?source=files>) at http://sourceforge.net/projects/scipy/files/

I am still seeing the memory leak, and it appears to be a little larger (25MB) each iteration. Here are the numbers I am seeing on my python process:

Iteration 1, sleep1: 207,716k
Iteration 1, sleep2: 212,112k

Iteration 2, sleep1: 236,488k
Iteration 2, sleep2: 237,160k

Iteration 3, sleep1: 261,168k
Iteration 3, sleep2: 261,264k

Iteration 4, sleep1: 285,044k
Iteration 4, sleep2: 285,724k

-Joe

From: numpy-discussion-bounces at scipy.org [mailto:numpy-discussion-bounces at scipy.org] On Behalf Of Ralf Gommers
Sent: Wednesday, January 29, 2014 6:16 AM
To: Discussion of Numerical Python
Subject: Re: [Numpy-discussion] scipy image processing memory leak in python 2.7?



On Tue, Jan 28, 2014 at 8:08 PM, Joseph McGlinchy <JMcGlinchy at esri.com<mailto:JMcGlinchy at esri.com>> wrote:
>>> scipy.version.version
'0.11.0b1'

>>> numpy.version.version
'1.6.1'

Is that, or any other, memory leaks present in these versions?

That memory leak isn't, it's only present in 0.13.0-0.13.2. I'm not aware of any other leaks that were fixed since 0.11.0 either, but still it would be worth checking if you see the same issue with current scipy master.
Ralf



-----Original Message-----
From: numpy-discussion-bounces at scipy.org<mailto:numpy-discussion-bounces at scipy.org> [mailto:numpy-discussion-bounces at scipy.org<mailto:numpy-discussion-bounces at scipy.org>] On Behalf Of Julian Taylor
Sent: Tuesday, January 28, 2014 10:58 AM
To: numpy-discussion at scipy.org<mailto:numpy-discussion at scipy.org>
Subject: Re: [Numpy-discussion] scipy image processing memory leak in python 2.7?

On 28.01.2014 19:44, Joseph McGlinchy wrote:
> Hi numpy list!
>
>
>
> I am trying to do some image processing on a number of images, 72 to
> be specific. I am seeing the python memory usage continually increase.

which version of scipy are you using?
there is a memory leak in ndimage.label in version 0.13 which will be fixed in 0.13.3 due soon.
see https://github.com/scipy/scipy/issues/3148


_______________________________________________
NumPy-Discussion mailing list
NumPy-Discussion at scipy.org<mailto:NumPy-Discussion at scipy.org>
http://mail.scipy.org/mailman/listinfo/numpy-discussion

_______________________________________________
NumPy-Discussion mailing list
NumPy-Discussion at scipy.org<mailto:NumPy-Discussion at scipy.org>
http://mail.scipy.org/mailman/listinfo/numpy-discussion

-------------- next part --------------
An HTML attachment was scrubbed...
URL: <http://mail.python.org/pipermail/numpy-discussion/attachments/20140129/775bc769/attachment.html>


More information about the NumPy-Discussion mailing list