[SciPy-user] FFT Problem

Nils Wagner nwagner at mecha.uni-stuttgart.de
Mon Jan 16 09:14:58 CST 2006


Frederick Ross wrote:
>I'm running the SVN builds from
>http://homepage.mac.com/fonnesbeck/mac/ on MacOS 10.4.4 on a 2005
>PowerBook G4, and have the following peculiar problem:
>
>  
>>>>import numpy
>>>>a = numpy.array([1, 3, 5, 4])
>>>>print numpy.ifft(numpy.fft(a))
>>>>        
>[ 1.+0.j  3.+0.j  5.+0.j  4.+0.j]
>  
>>>>import scipy
>>>>        
>Overwriting fft=<function fft at 0x783c70> from scipy.fftpack.basic
>(was <function fft at 0x6ae6b0> from numpy.dft.fftpack)
>Overwriting ifft=<function ifft at 0x783cb0> from scipy.fftpack.basic
>(was <function inverse_fft at 0x6ae6f0> from numpy.dft.fftpack)
>  
>>>>print scipy.ifft(scipy.fft(a))
>>>>        
>[ 1. +0.j  3.5-0.j  5. +0.j  3.5+0.j]
>
>Any suggestions on what might be causing this would be much appreciated.
>
>--
>Frederick Ross
>Graduate Fellow
>The Rockefeller University
>
>_______________________________________________
>SciPy-user mailing list
>SciPy-user at scipy.net
>http://www.scipy.net/mailman/listinfo/scipy-user
>  
I cannot reproduce your results. Note that there is no message starting
with  "Overwriting" on my system

>>> import numpy
>>> a = numpy.array([1, 3, 5, 4])
>>> print numpy.ifft(numpy.fft(a))
[ 1.+0.j  3.+0.j  5.+0.j  4.+0.j]
>>> import scipy
>>> print scipy.ifft(scipy.fft(a))
[ 1.+0.j  3.+0.j  5.+0.j  4.+0.j]
>>> scipy.__version__
'0.4.4.1556'
>>> numpy.__version__
'0.9.4.1914'

Nils



More information about the SciPy-user mailing list