[FieldTrip] new implementation of wltconvol (renamed to wavelet): specest

[Roemer van der Meij]

Hello everybody!

As of today, we have switched to a new implementation for the 
'wltconvol' method for frequency analysis, and it has also been renamed 
to 'wavelet'. For a while now we have been rewriting the low-level code 
in a different format, one that is more flexible to adapt in the future 
and that allows for the low-level algorithms to be downloaded in a 
separate module: /specest/. The changes will be on our ftp-server by 
tonight.

The switch to the new implementation brings about several changes. With 
respect to the changes are observable to the end-user, there are several 
FAQs created at our wiki.

For the end-user observable changes for 'wavelet' (formerly 
'wltconvol'), please have a look /here/ 
<http://fieldtrip.fcdonders.nl/faq/what_are_the_differences_between_the_old_and_the_new_implementation_of_wavelet_formerly_wltconvol_in_ft_freqanalyis>, 
and for a bigger description with respect to your output.freq please go 
/here/ 
<http://fieldtrip.fcdonders.nl/faq/why_does_my_output.freq_not_match_my_cfg.foi_when_using_wavelet_formerly_wltconvol_in_ft_freqanalyis>.


Because of these changes, we strongly advise to either use the new or 
the old implementation for your entire analysis project. The old 
implementation is still available by using 'wltconvol_old' as cfg.method 
(however, this code no longer being updated). You can call also call the 
function: ft_freqanalysis_old, which is the old interface-function.

If you are interested, you can track the progress of the /specest 
/module by going /here/ 
<http://fieldtrip.fcdonders.nl/development/specest>. As soon as the 
other low-level functions are ready and implemented, another e-mail will 
be sent to the mailing-list.

If anything is unclear, or if there are any bugs that have slipped 
through our fingers, please send an e-mail to the mailing-list and/or 
report a bug in our /Bugzilla bug tracking system 
<http://fieldtrip.fcdonders.nl/bugzilla>/.

Kind regards,

Roemer van der Meij


-- 
Roemer van der Meij MSc
PhD student
Donders Institute for Brain, Cognition and Behaviour
Centre for Cognition
P.O. Box 9104
6500 HE Nijmegen
The Netherlands
Tel: +31(0)24 3655932
E-mail: r.vandermeij at donders.ru.nl
-------------- next part --------------
An HTML attachment was scrubbed...
URL: <http://mailman.science.ru.nl/pipermail/fieldtrip/attachments/20110106/114381f7/attachment.html>