[Top][All Lists]

[Date Prev][Date Next][Thread Prev][Thread Next][Date Index][Thread Index]

[Octave-patch-tracker] [patch #8864] add smooth3

From: Markus Mützel
Subject: [Octave-patch-tracker] [patch #8864] add smooth3
Date: Mon, 18 Jan 2016 12:05:04 +0000
User-agent: Mozilla/5.0 (Windows NT 6.1; Win64; x64; rv:44.0) Gecko/20100101 Firefox/44.0


                 Summary: add smooth3
                 Project: GNU Octave
            Submitted by: mmuetzel
            Submitted on: Mo 18 Jan 2016 12:05:03 GMT
                Category: None
                Priority: 5 - Normal
                  Status: None
                 Privacy: Public
             Assigned to: None
        Originator Email: 
             Open/Closed: Open
         Discussion Lock: Any



This is a patch to add the function smooth3 to Octave.

I did some tests to compare the results in Matlab and Octave. But the results
differ slightly. The deviations are up to 5.5*eps in one example (see
attachments smooth3_test.m and smooth3_data.mat). They might be higher in
other cases.

I am not quite sure where the small deviations come from. Maybe Matlab and
Octave use different ways to save to and load data from mat-files. Maybe there
are differences in the implementation of some functions that are used (could
be "exp" and "convn"). Or they have a completely different reason...

I can test this only on Windows and do not know whether the results are
different on other platforms.

Are these small deviations an issue?

I do not know how to implement these manual tests as automated tests at the
end of the .m-file. Is it possible to read from .mat-files in the test?

Meanwhile, I added some tests that assert the size of the resulting matrix but
not really its content.


File Attachments:

Date: Mo 18 Jan 2016 12:05:03 GMT  Name: smooth3.m  Size: 8kB   By: mmuetzel

Date: Mo 18 Jan 2016 12:05:03 GMT  Name: smooth3_data.mat  Size: 45kB   By:

Date: Mo 18 Jan 2016 12:05:03 GMT  Name: smooth3_test.m  Size: 611B   By:



Reply to this item at:


  Nachricht gesendet von/durch Savannah

reply via email to

[Prev in Thread] Current Thread [Next in Thread]