return true if (infeasibility, objective) acceptable, false otherwise

Lagrangian gradient split in two parts: objective contribution and constraints' contribution

View the bytes written by snprintf into `buffer`. snprintf returns the would-be length (excluding the null terminator) or a negative value on error, s

check if the trial iterate is inside the current funnel */

check acceptability wrt current point

append leading space, `value`, then pad with spaces so the field occupies `width` glyphs

go a fraction along the direction by finding an acceptable step length returns true upon success, false upon failure

build HiGHS' lower-triangular CSC Hessian (model.hessian_) and the sorting permutation from the COO arrays already present in hessian_row_indices/hess

build HiGHS' CSC Jacobian (a_matrix_) and the sorting permutation from the COO arrays already present in jacobian_row_indices/jacobian_column_indices.

build BQPD's "la" sparsity array (gradients_sparsity) and the sorting permutation from the COO Jacobian already present in jacobian_row_indices/jacobi

performs Cholesky factorization of a symmetric positive definite matrix A A = U^T U or A = L L^T returns true upon success, false upon failure

protected member functions

flush buffer to the logger callback

getters

the trial iterate is accepted by the constraint relaxation strategy or if the step is small and we cannot switch to solving the feasibility problem

performs one of the matrix-matrix operations C := alpha op(A) op(B) + beta C, where op(X) = X or op(X) = X^T

x := alpha x