comp.lang.idl-pvwave archive: archive » Re: failed matrix inversion returns input-- interesting

Home » Public Forums » archive » Re: failed matrix inversion returns input-- interesting

Show: Today's Messages :: Show Polls :: Message Navigator
E-mail to friend

Re: failed matrix inversion returns input-- interesting [message #49302]

Fri, 14 July 2006 05:23

news.verizon.net
Messages: 47
Registered: August 2003

Member

> Is there any reason in creation why IDL simply copies the input into
> the output if it cannot do the inversion? That seems, well, malevolent,
> unless I'm missing something.

One reason you might want this behaviour is to preserve the original
array, in cases where the code overwrites the inverse

IDL> testm = invert(testm, status)
IDL> if status EQ 1 then print,'Singular Matrix ', testm

If INVERT returned, say, NaN values or an undefined variable (the only
reasonable alternatives I think), then you would have lost the original
array

Incidentally, the LAPACK (LU Decomposition) invert routine LA_INVERT in
IDL has the same behavior. ---Wayne

Report message to a moderator

[Message index]

		Re: failed matrix inversion returns input-- interesting By: news.verizon.net on Fri, 14 July 2006 05:23
		Re: failed matrix inversion returns input-- interesting By: mmeron on Fri, 14 July 2006 00:30
		Re: failed matrix inversion returns input-- interesting By: Bringfried Stecklum on Fri, 14 July 2006 00:14
		Re: failed matrix inversion returns input-- interesting By: marc schellens[1] on Fri, 14 July 2006 00:09

Previous Topic:	Re: IDL 6.3 for Mac OS X on Intel Now Available
Next Topic:	Re: IDL and Macs. Speed is not only about squared roots

-=] Back to Top [=-

[ Syndicate this forum (XML) ] [

] [

]

Current Time: Fri Oct 10 10:28:38 PDT 2025

Total time taken to generate the page: 0.48443 seconds