[Scipy-tickets] [SciPy] #1354: ffts are extremely memory hungry (memory leak?)

SciPy Trac scipy-tickets@scipy....
Wed Jan 12 07:28:00 CST 2011


#1354: ffts are extremely memory hungry (memory leak?)
---------------------------+------------------------------------------------
 Reporter:  bubla          |       Owner:  somebody
     Type:  defect         |      Status:  new     
 Priority:  normal         |   Milestone:  0.9.0   
Component:  scipy.fftpack  |     Version:  0.8.0   
 Keywords:                 |  
---------------------------+------------------------------------------------
 Hello, as stated in the summary, something needs to be done with the Scipy
 FFT implementation.

 Consider this example:

 import scipy as sp
 import scipy.fftpack
 import numpy as np
 import numpy.fft

 # Now watch the system's resource usage
 a = sp.rand(300,300,300)
 sp.fftpack.fftn(a)
 # Still lots of memory occupied
 np.fft.fftn(a)
 # Almost no additional memory occupied

 The point is that scipy uses much more resources and frees little, numpy
 is much better here.

-- 
Ticket URL: <http://projects.scipy.org/scipy/ticket/1354>
SciPy <http://www.scipy.org>
SciPy is open-source software for mathematics, science, and engineering.


More information about the Scipy-tickets mailing list