Golub and Van Loan
Matrix Computations 3rd edition
Johns Hopkins University Press
See section 4.7
- Mensagem original
De: Z F <[EMAIL PROTECTED]>
Para: [EMAIL PROTECTED]
Enviadas: Quinta-feira, 1 de Novembro de 2007 3:06:54
Assunto: Toeplitz solver
Hello everybody,
Could someone sugge
We could try to gslfy sparse-blas
http://math.nist.gov/spblas/
This weekend I might try to have a look on this.
Any suggestions?
--- Andrew Makhorin <[EMAIL PROTECTED]> escreveu:
> Hi,
>
> I am wondering if it is planned to eventually
> provide GSL with general
> sparse matrix routines (e.g. li
that?
>
> Thank you
>
> Regards,
>
> Ewen Carcreff.
>
> > De: Paulo Jabardo <[EMAIL PROTECTED]>
> > A: help-gsl@gnu.org
> > Objet: Re: [Help-gsl] svd complex routine
> > Date: Wed, 18 Apr 2007 14:27:48 -0300 (ART)
>
> > I don't
I don't know about gsl in this case but you can try
lapack - www.netlib.org/lapack. It has subroutines to
do that.
Paulo
--- Ewen Carcreff <[EMAIL PROTECTED]> escreveu:
> Hello,
>
> I am working on a C program that have to use a
> complex svd algorithm. The
> GSL does not provide such a function
andard was finished and the specification
wasn't tested at the time.
I have no experience with that...
[1] Josuttis, Nicolai, "The C++ Standard Library"
Paulo Jabardo
__
Fale com seus amigos de graça com o novo Yahoo! Messenger
http
han a simple math parser because
applications tend to grow and so do their needs and in
such cases lua is perfect.
Paulo Jabardo
--- Jigal A <[EMAIL PROTECTED]> escreveu:
> TNT looks very tempting.
> However - I thought of a combination of GSL and TNT.
> GSL, after all, has a very l
What could be interesting is to develop some
guidelines on serialization of GSL objects. Any
feature would have a standard way to write the code to
a file (or stream) and a standard way to read it. Some
sort of ASCII file format could be developed and
objects could be written and read using this fi
I recommend using LAPACK (www.netlib.org/lapack) or
clapack (C interface).
The function you are looking for is
dpbtrf (cholesky decomposition)
dpbtrs (uses the cholesky decomposition to solve the
linear system)
It does a Cholesky decomposition for banded matrices
(any bandwidth).
If you ha
) (&simul);
gsl_deriv_central(&fun, x, h, &result, &abserr);
}
You shouldn't be afraid of the void*. It is the
mechanism that GSL and several other libraries to
provide some flexibility. I hope this example helps.
Paulo Jaba
If I understand you correctly, you need to write a
wrapper class. Since your function should be defined
(in GSL) as gsl_function structure which has two
fields:
a function pointer to a function that accepts a double
and a void* and returns a double and a second field
that is a void*.
The params f
If the data is available in a regular grid then this
is the best solution. If it is not...
There is a procedure developed by Akima. The method
can be used to interpolate irregularly spaced data. It
basically builds a triangular mesh on the data and
uses this mesh to interpolate the data (linear or
There are two interfaces to multiply the matrices. The
The one you are attempting to use is the low level
cblas interface. In this case you should use something
like
cblas_dgemm(
CblasRowMajor,
CblasNoTrans,
CblasNoTrans,
n, l, n, 1.0, A->data, A->tda, B->data,
B->t
ods
...
}
--- Paulo Jabardo <[EMAIL PROTECTED]> escreveu:
> I've never used these functions but if I understand
> correctly your problem, I think you need to declare
> your member function static. If it is not declared
> static, there is a hidden argument in the method (a
>
I've never used these functions but if I understand
correctly your problem, I think you need to declare
your member function static. If it is not declared
static, there is a hidden argument in the method (a
pointer to the object itself, accessible through
"this").
The problem is that you probably
The BLAS/LAPACK has a naming convention. The ge part
means GEneral matrix. Other possibilities are sy for
SYmmetric matrices (stored in standard matrix format).
Then there are he for HErmitian, gb for General
Banded, sp for Symmetric Packed matrix, and there are
several other possibilities. That's
If I understand you correctly, this is easy. I've never used the
multidimensional minimization routines but the same strategy is used throughout
gsl.
To specify a function to be minimized there is the type
gsl_multimin_function_fdf or
gsl_multimin_function
In these structs, the functions are sp
I have no idea of that but I recommend using lapack (while the new version for
gsl is not out). I think you are looking for function dgeev
(www.netlib.org/lapack):
SUBROUTINE DGEEV( JOBVL, JOBVR, N, A, LDA, WR, WI, VL, LDVL, VR,
$ LDVR, WORK, LWORK, INFO )
CHARACTE
Thanks! I will take a closer look and will implement this in the following days.
Paulo
- Mensagem original
De: Brian Gough <[EMAIL PROTECTED]>
Para: help-gsl@gnu.org
Cc: Paulo Jabardo <[EMAIL PROTECTED]>
Enviadas: Terça-feira, 5 de Dezembro de 2006 18:02:27
Assunto: Re: Re
That's exactly the one I tried to "reverse engineer" since Jacobi polynomial
are small modification of that.
Paulo
- Mensagem original
De: Brian Gough <[EMAIL PROTECTED]>
Para: help-gsl@gnu.org
Cc: Paulo Jabardo <[EMAIL PROTECTED]>
Enviadas: Segunda-feira,
point operations.
Does any one have any hints or references on that that I could use?
Thanks for any help.
Paulo Jabardo
___
Novidade no Yahoo! Mail: receba alertas de novas mensagens no seu celular.
Registre seu aparelho
don't understand what you mean by "see the output". If that is plotting a
graph, GSL doesn't have graphing capabilities. You should use some other
package. Write the output to a file and use gnuplot or a spreadsheet program to
visualize the results.
Paulo Jabardo
Hello there--
Did you link to the library? If you didn't you would get this error.
g++ -I... prog.cpp -lgsl -lgslcblas
You can also use the program gsl-config, something like
g++ `gsl-config --cflags --libs` prog.cpp -o prog
#include
#include
int main(){
double x;
x=gsl_pow_2(4);
22 matches
Mail list logo
