forum.alglib.net http://forum.alglib.net/ |
|
c++ rmatrixlu - possibly bug in reorder out argument http://forum.alglib.net/viewtopic.php?f=2&t=46 |
Page 1 of 1 |
Author: | marcosassis [ Thu Jul 29, 2010 7:30 pm ] |
Post subject: | c++ rmatrixlu - possibly bug in reorder out argument |
Hi! I'm new to alglib, but I liked very much it's style and I would like to use it in my research. I tried to use rmatrixlu from c++ version, (last release: 2.6.0), and the decomposition worked well, but the compacted permutation matrix came wrong all the times. Something like (for a matrix 6x6): [5 5 3 5 5 5]. The way I understood this format, it could not have repeated values. Did some one have the same problem? Is possible I'm doing something wrong? |
Author: | Sergey.Bochkanov [ Fri Jul 30, 2010 5:30 am ] |
Post subject: | Re: c++ rmatrixlu - possibly bug in reorder out argument |
No, it's alright with function. From its documentation: Code: * P = P0*P1*...*PK, K=min(M,N)-1, Pi - permutation matrix for I and Pivots[I] [5 5 3 5 5 5] is a sequence of elementary permutations: 0th with 5th, 1st with 5th, 2nd with 3rd, 3rd with 5th. 4th with 5th, 5th with 5th. |
Author: | marcosassis [ Fri Jul 30, 2010 1:33 pm ] |
Post subject: | Re: c++ rmatrixlu - possibly bug in reorder out argument |
Sorry for my mystake, I'm used to other format. Thanks a lot! It helped very much! |
Page 1 of 1 | All times are UTC |
Powered by phpBB © 2000, 2002, 2005, 2007 phpBB Group http://www.phpbb.com/ |