Back to the main page.
Bug 1778 - processed data on ftp differs slightly from newly processed data
| Status | CLOSED FIXED |
| Reported | 2012-10-10 14:57:00 +0200 |
| Modified | 2014-03-12 12:21:45 +0100 |
| Product: | FieldTrip |
| Component: | core |
| Version: | unspecified |
| Hardware: | PC |
| Operating System: | Windows |
| Importance: | P3 normal |
| Assigned to: | Jörn M. Horschig |
| URL: | |
| Tags: | |
| Depends on: | 1151 |
| Blocks: | |
| See also: | http://bugzilla.fcdonders.nl/show_bug.cgi?id=1420 |
Jörn M. Horschig - 2012-10-10 14:57:50 +0200
probably due to some change in preprocessing/filtering/whatever question: replace ftp versions with recent versions? example can be found in test_tutorial_clusterpermutationtimelock.m but also in test_ft_freqanalysis, see bug 1420
Robert Oostenveld - 2012-10-10 19:34:16 +0200
(In reply to comment #0) Yes, please update the ftp data.
Jörn M. Horschig - 2013-03-12 10:49:21 +0100
let's wait until Diego finished the issue with the preprocessed data, first
Jörn M. Horschig - 2013-03-20 16:04:15 +0100
test_ft_freqanalysis: btigrad248, ctf64, itab28, itab28_old fail because of different grad-information, related to bug 1151 itab153,= fails because of one(!) small power value difference: K>> freq.powspctrm(192) ans = 5.1120e-30 K>> datanew.powspctrm(192) ans = 2.6072e-34 note that the mean power is 10^3 --> writeflag = 1 and overwrite 'em all
Jörn M. Horschig - 2013-03-20 16:26:35 +0100
test_ft_timelockanalysis: btigrad248, ctf64, itab28, itab28_old fail because of different grad-information, related to bug 1151 --> writeflag = 1 and overwrite 'em all
Jörn M. Horschig - 2013-03-25 17:04:17 +0100
and it's not fixed somehow... gonna look into this later again
Jörn M. Horschig - 2013-03-28 13:44:24 +0100
test_ft_freqanalysis runs on my windows machine and on a mentat using matlab20099b without problems. Robert, do you know what matlab version the cron job is using?