[Numpy-discussion] Normalization of ifft
Thu Mar 26 22:35:31 CDT 2009
> what you say is of course correct, but I am wondering if there is a
> mistake in the user guide (p. 180 of
> http://numpy.scipy.org/numpybook.pdf): according to the expressions in
> the user guide, both fft and ifft are not normalized. The
> implementation if ifft, on the other hand, has the additional 1/n
> factor, consistent with the online documentation.
You are looking at Travis Oliphant's book Guide to NumPy, last updated
2006. The routine docstrings (the "help" text) are now maintained by
the community at docs.scipy.org. They get synced into the source
repository relatively often. They are also the sources to the routine
docs presented in the NumPy Reference Guide, also available at that
site. Travis has freed his original book and large parts of it (e.g.,
the C API docs) are now being incorporated into the
actively-maintained manuals at docs.scipy.org. Please go there for
the latest docs. You'll find that the fft section gives the 1/n
formula when discussing ifft.
I can see where Lutz got the impression that Guide to Numpy was the
doc to read. The descriptions of books on both numpy.scipy.org and
docs.scipy.org do give that impression. But, Guide is "mature"
because its scope was (necessarily) limited. At this point the
Reference Guide, while not complete because of its more-ambitious
scope, has every docstring in Guide to Numpy, and has substantially
more complete and accurate pages for a large number of functions, and
much additional text that Guide to Numpy does not have. I haven't
checked in detail but much of the rest of Guide to Numpy is now
included in the Reference Guide. Would it be ok to put some words on
both sites to the effect that the RG is the place to go for routine,
class, and module docs, or (possibly) just the place to go, period?
I don't want to downplay Travis's contribution; Guide was *very*
useful and it lives on in the work descended from it. But, I think
the audience that should read it first is relatively limited now.
More information about the Numpy-discussion