search

Guokr1991 / addi

Automatically exported from code.google.com/p/addi
0 stars 0 forks source link

Wrong results with the command lowertriangle #61

Open GoogleCodeExporter opened 9 years ago

GoogleCodeExporter commented 9 years ago 
The attached m-File shows the correct and the wrong result. The matrices A, L 
and U are from a German Wikipedia page. The matrix L-lowertriangle(A) should be 
a null matrix.

Addi version is 1.91, hardware is a Samsung Galaxy Note with Android 2.3

Original issue reported on code.google.com by f.br...@gmail.com on 10 Apr 2012 at 8:48

Attachments:

GoogleCodeExporter commented 9 years ago 
Can you try going to settings/preferences inside addi and enabling the 
"experimental interpreter"?  This should be solved via that.  That will be the 
default behavior in the future.

Original comment by corbi...@gmail.com on 15 Apr 2012 at 3:33

GoogleCodeExporter commented 9 years ago

Original comment by corbi...@gmail.com on 15 Apr 2012 at 3:35

GoogleCodeExporter commented 9 years ago 
Thanks for your help. Unfortunatly in the preferences I find only the 
possibility to enable a custom keyboard. There is no possibility to enable an 
"experimental interpreter". My exact Android version is 2.3.6 and the Galaxy 
Note kernel version is N7000XXLC1.

Original comment by f.br...@gmail.com on 16 Apr 2012 at 7:33

GoogleCodeExporter commented 9 years ago 
You need to update to the latest version of Addi to have this option.

Original comment by corbi...@gmail.com on 16 Apr 2012 at 5:11

GoogleCodeExporter commented 9 years ago 
After an update to Addi version 1.95 and with "experimental interpreter" 
enabled my m-file stopps with the following message:
error: 'lowertriangle' undefined near line 5, column 3
error: called from:
error: /mnt/sdcard/...../test_addi_03.m at line 5, column 2
Looks like the command lowertriangle is not available with the "experimental 
interpreter".

Original comment by f.br...@gmail.com on 16 Apr 2012 at 9:40

GoogleCodeExporter commented 9 years ago 
With the "experimental interpreter" also the command uppertriangle is not 
available. Without the "experimental interpreter" the lower triangular factor 
of a matrix A can be evaluated with the expression A*inv(uppertriangle(A)). 
This is my avoidance for the error with the command lowertriangle.

Original comment by f.br...@gmail.com on 16 Apr 2012 at 10:26

GoogleCodeExporter commented 9 years ago 
http://www.math.utah.edu/docs/info/octave_24.html  see this.

Original comment by corbi...@gmail.com on 17 Apr 2012 at 4:43

GoogleCodeExporter commented 9 years ago 
[deleted comment]
GoogleCodeExporter commented 9 years ago 
In your link I find only the commands tril and triu which extract parts of a 
matrix but not the commands lowertriangle and uppertriangle as expected. My 
interpretation of lowertriangle and uppertriangle is that they should evaluate 
the triangular factors of a LU decomposition. At least with uppertriangle I 
really get the triangualar factor U of my test matrix A. Is my interpretation 
of the commands lowertriangle and uppertriangle correct?

Original comment by f.br...@gmail.com on 17 Apr 2012 at 10:51

GoogleCodeExporter commented 9 years ago 
Can you provide a link to the Octave or Matlab function you want emulated?  I 
am trying to move towards matlab/octave functionality.  I think you have 
latched onto a function that will no longer be supported.  I think you want 
[l,u,p]=lu(A).  http://www.delorie.com/gnu/docs/octave/octave_145.html

Original comment by corbi...@gmail.com on 18 Apr 2012 at 5:42

GoogleCodeExporter commented 9 years ago 
Ok, I accept that lowertriangle and uppertriangle are no longer supported. I 
appreciate very much your support. Thanks once more.
By the way the function [l,u,p]=lu(A) generates an wrong permutation matrix p 
in addi. Running the same function with Octave on my notebook the matrix p is 
correct.

Original comment by f.br...@gmail.com on 18 Apr 2012 at 8:16

GoogleCodeExporter commented 9 years ago 
Are you talking about the default or experimental interpreter when you say you 
got the wrong results?

Original comment by corbi...@gmail.com on 18 Apr 2012 at 3:01

GoogleCodeExporter commented 9 years ago 
The wrong results for the lu function are with the default interpreter; the 
experimental interpreter runs lu OK.

Original comment by f.br...@gmail.com on 18 Apr 2012 at 6:54

GoogleCodeExporter commented 9 years ago 
Ok good.  The experimental will be the default soonish.

Original comment by corbi...@gmail.com on 18 Apr 2012 at 7:21