unconstrained tool, gradient and Hessian sparse matrix
[G,H,ICNH,IRNH] = ugrsh(X)
real vector, variables of the problem.
real vector, gradient of the objective function evaluated at X.
real vector, value of the Hessian matrix of the objective function evaluated at X. H(i) gives the value of the nonzero in row IRNH(i) and column ICNH(i). Only the upper triangular part of the Hessian is stored.
an array which gives the column indices of the nonzeros of the Hessian matrix of the objective function evaluated at X.
an array which gives the row indices of the nonzeros of the Hessian matrix of the objective function evaluated at X.
[G,H,ICNH,IRNH] = ugrsh(X)
computes the
gradientG
and the Hessian matrix of a group
partially separable function. The upper triangle of the Hessian is
stored in coordinate form, i.e., the entry H(i)
has row index IRNH(i)
and column index
ICNH(i)
.
Serge Steer, INRIA
Based on CUTEr authored by
Nicholas I.M. Gould - n.gould@rl.ac.uk - RAL
Dominique Orban - orban@ece.northwestern.edu - Northwestern
Philippe L. Toint - Philippe.Toint@fundp.ac.be - FUNDP
see http://hsl.rl.ac.uk/cuter-www