Hi, I am planning to do that, but my data matrix is pretty large (say 100000 x 10), packing and unpacking for each minimization will be expensive. -Babai
On Tue, Feb 7, 2012 at 2:16 PM, Rhys Ulerich <rhys.uler...@gmail.com> wrote: >> I am trying to write a minimizer for a function that takes a data >> matrix (N x p) and a weight vector (P-dimensional) and a class table >> vector Y (N -dim). The Multidimensional Minimizer in GSL only allows a >> vector data type for the functions gsl_multimin_function and >> gsl_multimin_function_fdf. The two-dimensional paraboloid example >> shows how to pass a single vector >> of data (x,y). In my case I have a matrix. Is it possible to do such >> minimization with gsl ? > > Can you pack your matrix, weight vector, and vector into a contiguous > region of memory and then use a gsl_vector_view > (http://www.gnu.org/software/gsl/manual/html_node/Vector-views.html) > to view that memory as a single "vector" to accomplish what you want? > > - Rhys