Back to the main page.
Bug 2040 - nansum fallback behaves incompatible with the nansum mex file
| Status | CLOSED FIXED |
| Reported | 2013-03-09 23:18:00 +0100 |
| Modified | 2019-08-10 12:02:47 +0200 |
| Product: | FieldTrip |
| Component: | core |
| Version: | unspecified |
| Hardware: | PC |
| Operating System: | Mac OS |
| Importance: | P3 normal |
| Assigned to: | Martin Vinck |
| URL: | |
| Tags: | |
| Depends on: | |
| Blocks: | |
| See also: |
Martin Vinck - 2013-03-09 23:18:00 +0100
it defaults to the first dimension
Martin Vinck - 2013-03-09 23:40:42 +0100
committed new nansum.m which behaves similarly as the mex file and matlab's nansum
Robert Oostenveld - 2013-03-10 00:01:09 +0100
to be precise, the drop in replacement function nansum.m behaves differently from the mex version and from the stats toolbox version. TODO: make a test script that demonstrates that the present implementation is correct.
Robert Oostenveld - 2013-03-10 12:06:47 +0100
Created attachment 435 proposed test script (In reply to comment #2) I have made a test script. Note that it is not on svn yet, but in my local git copy. And it fails as expected, since I have not updated my fieldtrip copy yet. laptop> git commit [bug2040 27ca61c] enhancement - implemented test script for http://bugzilla.fcdonders.nl/show_bug.cgi?id=2040 0 files changed, 0 insertions(+), 0 deletions(-) create mode 100644 test/test_bug2040.m
Martin Vinck - 2013-03-13 15:09:03 +0100
I have used the test script and the fallback behaves the same as the mex file