Skip to content

Commit

Permalink
Update documentation and NEWS.md
Browse files Browse the repository at this point in the history
  • Loading branch information
giordano committed Dec 17, 2016
1 parent 648bfb8 commit 99d4fe6
Show file tree
Hide file tree
Showing 2 changed files with 25 additions and 4 deletions.
16 changes: 16 additions & 0 deletions NEWS.md
Original file line number Diff line number Diff line change
@@ -1,6 +1,22 @@
History of LombScargle.jl
=========================

v0.3.0 (201?-??-??)
-------------------

### Breaking Changes

* Support for Julia 0.4 was dropped.

### New Features

* The fast method now is a bit faster. The fast Fourier transform computed
internally with FFTW library can take advantage of multi-threading (call
`FFTW.set_num_threads(n)` to use `n` threads) in order to speed-up
computation. However, please note that the running time will not scale as
`1/n` because computation of the FFT is only a part of the algorithm. The
function is also less memory-eager than before.

v0.2.0 (2016-12-07)
-------------------

Expand Down
13 changes: 9 additions & 4 deletions docs/index.rst
Original file line number Diff line number Diff line change
Expand Up @@ -221,10 +221,16 @@ datapoints, the more accurate the approximation.
.. Note::

This method internally performs a `Fast Fourier Transform
<https://en.wikipedia.org/wiki/Fast_Fourier_transform>`_ to compute some
quantities, but it is in no way equivalento to conventional Fourier
<https://en.wikipedia.org/wiki/Fast_Fourier_transform>`_ (FFT) to compute
some quantities, but it is in no way equivalento to conventional Fourier
periodogram analysis.

``LombScargle.jl`` use FFTW functions to compute the FFT. You can speed-up
this task by using multi-threading: call ``FFTW.set_num_threads(n)`` to use
:math:`n` threads. However, please note that the running time will not scale
as :math:`1/n` because computation of the FFT is only a part of the
algorithm.

The only prerequisite in order to be able to employ this fast method is to
provide a ``frequencies`` vector as a ``Range`` object, which ensures that the
frequency grid is perfectly evenly spaced. This is the default, since
Expand Down Expand Up @@ -257,8 +263,7 @@ can use the fast method:

* by default if the length of the output frequency grid is larger than 200
points
* if the frequency grid has 200 points or less and you explicitely request the
method with the ``fast=true`` keyword
* in any case with the ``fast=true`` keyword

Setting ``fast=false`` always ensures you that this method will not be used,
instead ``fast=true`` actually enables it only if ``frequencies`` is a
Expand Down

0 comments on commit 99d4fe6

Please sign in to comment.