Back to the main page.
Bug 2059 - cfg.channel in ft_topoplotER
Status | ASSIGNED |
Reported | 2013-03-15 22:43:00 +0100 |
Modified | 2013-03-18 13:30:40 +0100 |
Product: | FieldTrip |
Component: | plotting |
Version: | unspecified |
Hardware: | PC |
Operating System: | Windows |
Importance: | P3 normal |
Assigned to: | Robert Oostenveld |
URL: | |
Tags: | |
Depends on: | |
Blocks: | |
See also: | http://bugzilla.fcdonders.nl/show_bug.cgi?id=1977 |
Anne van Hoogmoed - 2013-03-15 22:43:30 +0100
Created attachment 437 GA_loc_diff&lay_baby I've previously used cfg.channel to define which channels I wanted to have included in ft_topoplotER. This doesn't seem to work anymore, it just uses all channels in the layout. The script looks like this: cfg = []; cfg.channel = {'4','5','6','8','9','10','11','13','14','15','19','20','21','24','25','26'}; cfg.layout = lay_baby; cfg.xlim = [0.7 1.2]; cfg.zlim = [-6 6]; ft_topoplotER(cfg, GA_loc_diff); I've attachted the files. Thanks, Anne
Robert Oostenveld - 2013-03-16 08:07:22 +0100
thanks for the detailed bug report. I have confirmed the problem, with cfg = []; cfg.channel = {'4','5','6','8','9','10','11','13','14','15','19','20','21','24','25','26'}; cfg.layout = lay_baby; cfg.xlim = [0.7 1.2]; cfg.zlim = [-6 6]; cfg.channel = 'all'; figure ft_topoplotER(cfg, GA_loc_diff); cfg.channel = {'4','11','16','19','20','21'}; figure ft_topoplotER(cfg, GA_loc_diff); you would expect two different figures, but both are the same.
Robert Oostenveld - 2013-03-16 08:12:11 +0100
I identified the problem, but don't know yet how to solve it. For an unclear reason, the channelselection works differently in interactive and non-interactive mode. Probably this has been the case for a longer time. But recently we switched the interactive mode on by default, which explains that the figures now look different. @Anna, a workaround for the moment is to do cfg.interactive = 'no'
Robert Oostenveld - 2013-03-16 09:10:43 +0100
after some struggles with git and svn, I have updated the code The code change that solves this specific instance is in ft_prepare_layout. In the case of an input cfg.layout structure (as here), it would not make the selection of the channels. With other layout specifications (such as cfg.layout='CTF151.lay') it would. @Anne: The workaround is not needed any more. I have also added a test script that compares the figures. With cfg.interactive=yes/no the figure is now the same, with different channel selections the figure is now different. Please see http://code.google.com/p/fieldtrip/source/detail?r=7672 for low-level details. Due to the git struggles, the detailed commit log messages have disappeared. Still to do is to discuss in the FT meeting why there is some behavior in the topoplot_common function that depends on interactive=yes/no. That seems an explicit design choice, but I don't get it.
Jörn M. Horschig - 2013-03-18 12:41:43 +0100
similar issue for ft_multiplotTFR. Not sure how topoplot is calling ft_prepare_layout, but the fix does work with ft_multiplotTFR, see bug 2066. I mark this bug as blocking, because imho the same fix as used here should be applicable to multiplot
Jörn M. Horschig - 2013-03-18 12:42:40 +0100
(In reply to comment #4) Not sure how topoplot is calling ft_prepare_layout, but the fix does *not yet* work with ft_multiplotTFR And I need an edit function ;)
Jörn M. Horschig - 2013-03-18 13:29:52 +0100
*** Bug 2066 has been marked as a duplicate of this bug. ***