[SciPy-user] Rank-Revealing QR factorization of dense matrices in scipy

Robert Kern rkern at ucsd.edu
Tue Oct 12 09:57:20 CDT 2004


Nils Wagner wrote:
> Robert Kern wrote:
> 
>> Nils Wagner wrote:
>>
>>> Hi all,
>>>
>>> Has someone written a function for Rank-Revealing QR factorization of 
>>> dense matrices in scipy ?
>>
>>
>>
>> No. The relevant LAPACK function set is xGEQP3, I believe. 
>> Contributions are welcome.
>>
> BTW, a good source is
> 
> C. H. Bischof, G. Quintana-Orti
> Algorithm 782: Codes for Rank-Revealing QR Factorizations of Dense
> Matrices
> 
> ACM Transactions on Mathematical software, Vol.24 No. 2 pp. 254-257

The LAPACK functions are unencumbered by TOMS's non-commercial-use-only 
license. I'm fairly sure, though, that xGEQP3 is a  version of the 
TOMS782 code.

[snip]

> It would be nice to have a built-in function (e.g. rank(A)) for this 
> purpose.

Indeed. If you write one that's sufficiently robust, documented, and 
unit tested, I'll check it in. Until then, the svdvals solution should 
work just fine for you.

> Maybe linalg.orth can be used in this context.

No, that computes the full SVD, which is overkill.

-- 
Robert Kern
rkern at ucsd.edu

"In the fields of hell where the grass grows high
  Are the graves of dreams allowed to die."
   -- Richard Harter



More information about the SciPy-user mailing list