Re: manipulating sparse matrices

[Mike Miller]

On Sat, 15 Oct 2005, David Bateman wrote:

> nzmax is in 2.9, as the source-forge index was I believe based on a 
> 2.9.2 release, it should probably be updated soon as many things have 
> changed (eg statistics functions, etc). The 0bytes is a missing feature 
> in the octave-forge sparse functions as the bytes size in the whos 
> function I believe was added at 2.1.58, after the octave-forge sparse 
> stuff was written. Basically, it needs to implement a byte_size() 
> function to return the value that will be printed.. This also works in 
> 2.9.3

For the meantime, stuck as I am in the 2.1.71 world (because I'm 
developing code for users who can install 2.1.71), can I estimate nzmax by 
taking 8*nnz()?  It seems that full matrices use 8 bytes per element.  Or 
maybe it should be 3*8*nnz() because every sparse element requires three 
terms: row, column and value.


> As you were asking about cholesky factorizations previously, I've just 
> gotten sparse cholesky factorizations working with cholmod and will 
> commit this soon, I want to try to address the chol2inv, cholinv and inv 
> functions, and update the docs first though...

Excellent!

You guys are great.  Thanks.

I suppose that Octave 2.9.x will be released someday as Octave 3.0. 
Is that the plan?

Mike



-------------------------------------------------------------
Octave is freely available under the terms of the GNU GPL.

Octave's home on the web:  http://www.octave.org
How to fund new projects:  http://www.octave.org/funding.html
Subscription information:  http://www.octave.org/archive.html
-------------------------------------------------------------