Re: Octave/C++ matrix Inv() comparison

[CdeMills]

ionone wrote
> thank you for your thorough answer
> 
> What i'm trying to do is to implement a matlab algorithm in C++
> 
> starting from there, I just copied the matlab functions in C++. It's a 
> DSP algorithm. And with all the inverse functions i used, i have altered 
> outputs.
> 
> for exemple, if in the input i have the matrix : R = [1/1024;1/1024;1], 
> i expect, after several calculations to have something also growing 
> slowly like : 0.021 0.025 0.030, etc

Could you provide a few simple Octave statements to generate your test case
?

You may also have a look at [1], the multi-precision package. I implemented
a simple inversion algorithm using the idea found in [2]. The idea is to
create an augmented matrix with an identity matrix. The base matrix is then
reduced to the identity matrix using simple, linear transformations. The
same operations are applied on the identity matrix; when the base matrix is
reduced to identity, the identity matrix is transformed into the inverse of
A. This algorithm is implemented in @mp/private/mp_diag_adj

Regards

Pascal

[1] https://sourceforge.net/p/octave/multi-precision/ci/default/tree/
[2]
http://www.mathsisfun.com/algebra/matrix-inverse-row-operations-gauss-jordan.html



--
View this message in context: 
http://octave.1599824.n4.nabble.com/Octave-C-matrix-Inv-comparison-tp4655291p4655508.html
Sent from the Octave - Maintainers mailing list archive at Nabble.com.