Back to the main page.
Bug 2004 - ft_connectivityanalysis: improve support for source level frequency domain data
| Status | CLOSED FIXED |
| Reported | 2013-02-23 10:50:00 +0100 |
| Modified | 2016-11-29 09:04:26 +0100 |
| Product: | FieldTrip |
| Component: | core |
| Version: | unspecified |
| Hardware: | PC |
| Operating System: | Mac OS |
| Importance: | P3 normal |
| Assigned to: | Jan-Mathijs Schoffelen |
| URL: | |
| Tags: | |
| Depends on: | |
| Blocks: | 2468 |
| See also: |
Jan-Mathijs Schoffelen - 2013-02-23 10:50:24 +0100
working with Alex Backus on a tighter integration with source level data obtained with 'pcc' as method. -inventorise whether it works: *which methods? *why doesn't it work for some methods? *fix it.
Jan-Mathijs Schoffelen - 2013-02-23 12:59:20 +0100
bash-3.2$ svn commit -m "enhancement - added testfunction for bug2004" test_bug2004.m Adding test_bug2004.m Transmitting file data . Committed revision 7535. I also uploaded some data (provided by Alex) to /home/common/matlab/fieldtrip/data/test/ (bug2004.mat)
Jan-Mathijs Schoffelen - 2013-02-23 14:27:35 +0100
bash-3.2$ svn commit -m "enhancement - built in support for powcorr_ortho for source level data" ft_connectivityanalysis.m connectivity/ft_connectivity_powcorr_ortho.m test/test_bug2004.m Sending connectivity/ft_connectivity_powcorr_ortho.m Sending ft_connectivityanalysis.m Sending test/test_bug2004.m Transmitting file data ... Committed revision 7536. @Stan: I changed the behavior of ft_connectivity_powcorr_ortho, so that the two correlation coefficients (orthogonalising w.r.t. one or the other) are averaged. Perhaps we may discuss the necessity for this to begin with.
Jan-Mathijs Schoffelen - 2015-01-08 13:07:26 +0100
At the moment the test script seems to run through just fine. it checks for coh powcorr amplcorr and powcorr_ortho. Closing for now.
Stan van Pelt - 2015-01-08 13:08:27 +0100
OK
Jan-Mathijs Schoffelen - 2015-05-07 12:06:42 +0200
With the new source representation the test function fails in univariate2bivariate, where the 'pow' does not exist (and should probably be computed from the mom :-) ).