release candidate 1.2.90

release candidate 1.2.90

[Brian Gough]

http://www.getafile.com/cgi-bin/merlot/get/network-theory/gsl/gsl-1.2.90.tar.gz

Please test it out and report the results to
gsl-discuss@sources.redhat.com

Brian


** Changed interface for gsl_sf_coupling_6j...(...). The old functions
actually calculated 6j for a permutation of the arguments (that
related to Racah W). This was incorrect and not consistent with
the documentation. The new versions calculate < {a,b,c}, {d,e,f} >,
as stated in the documentation. The old versions are still available
as gsl_sf_coupling_6j_INCORRECT...(...), though they are deprecated
and will be removed at some point in the future.
   
** Fixed compilation problems with gcc -ansi and other ANSI compilers.

** Fixed uninitialized memory access in the Niederreiter quasi-random
number generator.

** Fixed the eigenvalue routines to prevent an infinite loop for Inf
or NaN entries in matrix.

** Fixed a bug in the multifit and multiroots allocation routines
which cause them to fail to report some out of memory conditions.

** Fixed a bug in the seeding for the random number generator
gsl_rng_taus2 which affected a small number of seeds.

** Modified the complex householder transforms to avoid division by
zero, which could cause NaNs to be returned by the gsl_eigen_hermv
eigenvalue decomposition.

** The Nelder-Mead simplex algorithm for multidimensional 
minimisation has been added.

** The random number distributions now include the Dirichlet and
Multinomial distributions.

** Added a new function gsl_fcmp for approximate comparison of
floating point numbers using Knuth's algorithm.

** Added new functions gsl_ldexp and gsl_frexp as portable
alternatives to ldexp() and frexp().

** Fixed a bug in gsl_linalg_bidiag_unpack_B which was returning
incorrect results for the superdiagonal.

** Fixed a bug in the acceptance condition for simulated annealing

** Ordinary differential equations can now be solved using a different
absolute error for each component with gsl_odeiv_control_scaled_new().

Re: release candidate 1.2.90

[Kenneth Geisshirt]

On Wed, 2002-12-11 at 19:07, Brian Gough wrote:

> Please test it out and report the results to
> gsl-discuss@sources.redhat.com

All tests pass on my Mandrake 9.0 computer at work (Pentium IV, gcc 3.2,
glibc 2.2.5).

Kneth
 

Re: release candidate 1.2.90

[Lukas Dobrek]

On Wed, Dec 11, 2002 at 06:07:02PM +0000, Brian Gough wrote:
> http://www.getafile.com/cgi-bin/merlot/get/network-theory/gsl/gsl-1.2.90.tar.gz
> 

with:  
dobrek:dobrek/gsl/gsl-1.2.90> gcc --version
2.95.4

on Axp ev6 with compilation flags:

gcc -c -mieee -mfp-rounding-mode=d -g -O2  conftest.c

make check dies in:

/scratch/boniek/dobrek/gsl/gsl-1.2.90/eigen

I got  nan insted of 0 in herm(10) test but this I thing is 
a known issue.

Appart of it everything looks fine.


Lukasz Dobrek

PS. Thanks for this library.
 
-- 
£ukasz Dobrek
   An optimist believes that we live in the best of all possible worlds.
   A pessimist is sure that this must be so.
http://www.pld-linux.org

Re: release candidate 1.2.90

[Erik Ø. Sørensen]

Compiled on Apple iBook (G3) with
CC=gcc2 CFLAGS="-Wno-long-double -g -O" ./configure --disable-shared,
and all tests passed.

gcc2 is 2.95.2.

Attempts with the default gcc, 3.1, failed with compiler error.
Attemps with -O2 failed when testing.

best regards,
Erik

--
Erik Ø. Sørensen,  <http://www.geocities.com/erik_oiolf/>.
phd student (economics), Norwegian School of Economics.
currently visiting Queen's University, Kingston, Ontario.

Wishart Distribution

[Adam Johansen]

Hello,

Unless I've been looking in the wrong place, the Wishart distribution
doesn't appear to be included in the gsl random number distributions. As
I'm probably going to have to implement one anyway, I was wondering
whether there would be any interest in including it within the gsl (I'd
like to know before I start as it will obviously effect my coding style
and choice of variable names, etc.). It seems likely that it will be
useful to at least some users as the conjugate distribution to the
precision matrix of a multivariate normal distribution.

Apologies if I've missed something which is already present, or it's been
decided not to include this distribution.

Thanks,
Adam

Re: release candidate 1.2.90

[Erik Ø. Sørensen]

I should have made it clear in my previous mail: This is on Mac OS 
X.2.2:

On torsdag, des 12, 2002, at 10:02 America/Montreal, Erik Ø. Sørensen 
wrote:

> Compiled on Apple iBook (G3) with
> CC=gcc2 CFLAGS="-Wno-long-double -g -O" ./configure --disable-shared,
> and all tests passed.
>
> gcc2 is 2.95.2.
>
> Attempts with the default gcc, 3.1, failed with compiler error.
> Attemps with -O2 failed when testing.
>
> best regards,
> Erik
>
> --
> Erik Ø. Sørensen,  <http://www.geocities.com/erik_oiolf/>.
> phd student (economics), Norwegian School of Economics.
> currently visiting Queen's University, Kingston, Ontario.
>
>
--
Erik Ø. Sørensen,  <http://www.geocities.com/erik_oiolf/>.
phd student (economics), Norwegian School of Economics.
currently visiting Queen's University, Kingston, Ontario.

Re: release candidate 1.2.90

[Giulio Bottazzi]

Brian Gough wrote:

>http://www.getafile.com/cgi-bin/merlot/get/network-theory/gsl/gsl-1.2.90.tar.gz
>
>Please test it out and report the results to
>gsl-discuss@sources.redhat.com
>
>Brian
>
>  
>
All tests passed on these:

1 -
CPU: Pentium4 - M 2.20GHz
OS: Linux (Debian 3.0)
compiler: gcc version 2.95.4 20011002 (Debian prerelease)
libc: glibc 2.2.5

2 -
CPU: AMD Athlon XP 1800+
OS: Linux (SuSE 8.1)
compiler: gcc version 3.2
libc: glibc 2.2.5

Giulio.

Re: release candidate 1.2.90 -- all checks passed on Compaq Alpha

[Slaven Peles]

> Brian Gough wrote:
> >http://www.getafile.com/cgi-bin/merlot/get/network-theory/gsl/gsl-1.2.90.t
> >ar.gz
> >
> >Please test it out and report the results to
> >gsl-discuss@sources.redhat.com
> >
> >Brian
>

Compiled successfully and all tests passed on Compaq Alpha.

Compaq C V6.3-025 on Compaq Tru64 UNIX V5.1 (Rev. 732)
Compiler Driver V6.3-026 (sys) cc Driver

Slaven

Re: release candidate 1.2.90

[Stoffel]

All test successfully passed on
Debian ppc unstable, TiBook1st, gcc2.95.4 and gcc3.2.2.
CFLAGS="-O2"

jops,
	Stoffel.

On Wed, Dec 11, 2002 at 06:07:02PM +0000, Brian Gough wrote:
> http://www.getafile.com/cgi-bin/merlot/get/network-theory/gsl/gsl-1.2.90.tar.gz
> 
> Please test it out and report the results to
> gsl-discuss@sources.redhat.com
> 
> Brian
-- 
Anything that is good and useful is made of chocolate.

Re: release candidate 1.2.90

[Dirk Eddelbuettel]

Debian packages for 1.2.90-1 were uploaded yesterday from an i386 built.
Packages for nine other of the architectures have already been auto-built by
now, one arch (mips) seems to lag behind.  All builds ran all tests.

Details are at http://buildd.debian.org/build.php?pkg=gsl

Dirk

-- 
Prediction is very difficult, especially about the future. 
				             -- Niels Bohr

Re: Wishart Distribution

[Brian Gough]

Adam Johansen writes:
 > Hello,
 >  Unless I've been looking in the wrong place, the Wishart
 > distribution doesn't appear to be included in the gsl random number
 > distributions. As I'm probably going to have to implement one
 > anyway, I was wondering whether there would be any interest in
 > including it within the gsl (I'd like to know before I start as it
 > will obviously effect my coding style and choice of variable names,
 > etc.). It seems likely that it will be useful to at least some
 > users as the conjugate distribution to the precision matrix of a
 > multivariate normal distribution.

There is a GSL Design document at http://sources.redhat.com/gsl/

The main issue with multivariate distributions is to factor out any
memory allocation/matrix decomposition since this only needs to be
done once, not for each sample.

Brian

Re: release candidate 1.2.90

[Pedro Gonnet]

all tests passed.

cheers
pedro


machine:
--------
Intel Pentium 4 CPU 2.00GHz
256kB cache
1GB RAM

OS:
---
Linux 2.4.9-21
Red Hat Linux 7.1

compiler
--------
gcc 3.2
CFLAGS="-O4 -fexpensive-optimizations -march=pentium4 -mfpmath=sse
-msse2"