Re: Singular matrix

[Siki Zoltan]

Dear Lev,

the \ operator works for singular and non  quadratic matrices. In this 
case it search for the minimum norm solution.

for example
inv(A)
ans =

   Inf   Inf
   Inf   Inf

pinv(A)
ans =

   0.10000   0.10000
   0.20000   0.20000

pinv(A) * B
ans =

   2.1000
   4.2000

For more details see
https://www.gnu.org/software/octave/doc/v4.0.1/Arithmetic-Ops.html

Regards
Zoltan

On Mon, 3 Oct 2016, Lev wrote:

> Octave 4.0.3
>
> I was playing with simple linear systems.
>
> A=[1,2;1,2]
> B=[10;11]
>
> > > A
> A =
>
>   1   2
>   1   2
>
> > > B
> B =
>
>   10
>   11
>
> > > A\B
> ans =
>
>   2.1000
>   4.2000
>
> Which is not the thing I expected. I expected some overflow, since this is 
> singular.
>
> Scilab answered this:
>
> matrix is close to singular or badly scaled. rcond =    0.0000D+00
> computing least squares solution. (see lsq).
>
>
> Do I miss something?
>
> --
> 73 de HA5OGL
> Op.: Levente
> _______________________________________________
> Help-octave mailing list
> address@hidden
> https://lists.gnu.org/mailman/listinfo/help-octave