GNU Scientific Library

[infradig]

I have started an FFI module for the GNU Scientific Library, see library/gsl.pl

As this is an extensive library it is barely conceivable that all of it will be covered (BLAS alone of course is huge, those who know it will be aware), unless some method of automated assimilation is used. Which is unlikely. So initially I will probably just add some of what I have been doing in the matrix branch, in order to contrast and compare.

[infradig]

Seems feasible...

$ tpl
?- use_module(library(gsl)).
   true.
?- gsl_matrix_calloc(10,10,M), gsl_matrix_set_identity(M), gsl_matrix_minmax(M,Min,Max), gsl_matrix_free(M).
   M = 109675795343088, Min = 0.0, Max = 1.0.
?- 

[infradig]

Added read/write...

$ tpl
?- use_module(library(gsl)).
   true.
?- open('samples/test_matrix_det.mat',read,S),mat_read(M,S,Rows,Cols),mat_write(M,stdout).
#36,6x6,0.0
64,8,36,0,88,77
53,54,17,97,72,74
22,96,63,30,48,31
71,21,94,50,63,61
1,42,45,55,96,52
14,99,87,80,51,38
   S = '<$stream>'(4), M = 94377575873728, Rows = 6, Cols = 6.
?- 
~/trealla (devel) $ cat samples/test_matrix_det.mat
#36,6x6,0.0
64,8,36,0,88,77
53,54,17,97,72,74
22,96,63,30,48,31
71,21,94,50,63,61
 1,42,45,55,96,52
14,99,87,80,51,38
#det=34637561460.0

[infradig]

Have mat_lup_det working:

$ tpl -g main,halt samples/test_matrix_det.pl 
34637561460.0


Which agrees with the *matrix* branch.