Re: Golub-Reinsch SVD algorithm by Harold Norman + soft copy avaliable

From: Peter Spellucci (spellucci_at_fb04373.mathematik.tu-darmstadt.de)
Date: 02/21/05 

Date: Mon, 21 Feb 2005 16:47:34 +0000 (UTC)

In article <trqrwggn3q4y@legacy>,
 sathish@au-kbc.org (Sathish Viswanathan) writes:
>Hi friends..
> I am using the SVDcmp.c file written in "Numerical recipies in C"
>book. The code is actually based on Golub-Reinsch algorithm and it is
>difficult to get the idea about the algorithm.. Is it possible to get
>the book containing the algorithm in online or atleast the a document
>about the algorithm?..
>
>Hope anyone of you can help me..
>
>thanx and regards
>
>V.Sathish
>MS by Research
>AUKBC Research centre
>Chennai
>India -44
>Phone no: 91-44-22234885, 22232711, 22231034
>

the idea of the algorithm is well described in
Golub, G.H.; van Loan, Ch. : Matrix Computations (3ed ed.) John Hopkins Press
the original algorithm , with background is in
Wilkinson j.H. ; Reinsch, Chr.: Handbook for Automatic Computation Vol II
Linear Algebra. Grundlehren der Math. Wiss. 186 (Springer) (in English).
Warning: in the hope to do something good the authors of NR changed the
termination criterion to a "machine independent one" which quite often fails
since it is to restrictive (somethng like if ( x+y==x ) { ....} )
better go back to the original version.
hth
peter

Relevant Pages

  • Re: Can EXPTIME and NP be separated via diagonalization?
    ... algorithm that finds an algorithm that solves SAT. ... To multitask over all these algorithms, just zigzag your way through this ... grid, as in the proof that the rational numbers are countable. ... and beyond my knowledge without to modify the original algorithm. ...
    (comp.theory)
  • Re: Basically a sieve method, relation to quantum
    ... >> The original algorithm in my program, will, my current analysis ... > I'm not sure why that's astounding. ... Yeah but my algorithm does it in polynomial time. ...
    (sci.math)
  • Re: Basically a sieve method, relation to quantum
    ... >> The original algorithm in my program, will, my current analysis ... > I'm not sure why that's astounding. ... Yeah but my algorithm does it in polynomial time. ...
    (sci.crypt)
  • Re: Matrix equation
    ... > definite matrix through Cholesky decomp? ... According to Golubs and van Loans "Matrix Computations 2nd ed." Cholesky ... Also if Alan Tan didn't do any ... Is this a _new_ algorithm? ...
    (comp.dsp)
  • Golub-Reinsch SVD algorithm by Harold Norman + soft copy avaliable
    ... I am using the SVDcmp.c file written in "Numerical recipies in C" ... The code is actually based on Golub-Reinsch algorithm and it is ... AUKBC Research centre ...
    (sci.math.num-analysis)