470,647 Members | 1,219 Online
Bytes | Developer Community
New Post

Home Posts Topics Members FAQ

Post your question to a community of 470,647 developers. It's quick & easy.

FFT with Python

Hi All,

Did anyone compute FFT - Fast Fourier Trans using Python ? I am looking for a simple Python package which can do spectral analysis for me. Can someone help me get such a package ?
Thanks

Regards,
Satish Chimakurthi
Jul 18 '05 #1
3 2309
Satish Chimakurthi <sk*****@engr.uky.edu> pisze:
Did anyone compute FFT - Fast Fourier Trans using Python ? I am
looking for a simple Python package which can do spectral analysis for
me. Can someone help me get such a package ?


Numeric has it.

--
Jarek Zgoda
http://jpa.berlios.de/
Jul 18 '05 #2
On Tue, 22 Jun 2004 21:05:33 +0000, Jarek Zgoda wrote:
Satish Chimakurthi <sk*****@engr.uky.edu> pisze:
Did anyone compute FFT - Fast Fourier Trans using Python ? I am
looking for a simple Python package which can do spectral analysis for
me. Can someone help me get such a package ?


Numeric has it.


Of course you probably want to use numarray instead, which is newer and
has further development.

Jeremy

Jul 18 '05 #3
Jeremy Sanders wrote:
Of course you probably want to use numarray instead, which is newer and
has further development.


Unless speed for small arrays matters, case in which you stick with Numeric
until the Numarray constructors and some other functions get rewritten in C
(post 1.0). Numarray works great (and is better documented, cleaner, etc) for
the 'few large arrays' case. It has problems for the 'many small arrays' type
of code. That's why SciPy hasn't switched to Numarray.

Numarray is being written by the Hubble crowd, so it makes sense that they'd
optimize their usage cases first (astronomy images are the 'few big arrays'
kind of code).

Just a small precision.

And for the OP, scipy wraps fftw for you, as well as including a raft of other
signal-processing oriented things. And it makes sure to do use ATLAS build of
lapack.

hth,

f
Jul 18 '05 #4

This discussion thread is closed

Replies have been disabled for this discussion.

By using this site, you agree to our Privacy Policy and Terms of Use.