ConvexQP_Solvers (functions)¶

siconos.numerics.convexQP_ADMM(ConvexQP *problem, array_like (np.float64, 1D)z, array_like (np.float64, 1D)w, array_like (np.float64, 1D)xi, array_like (np.float64, 1D)u, int *info, SolverOptions *options) → None[source]¶

siconos.numerics.convexQP_ADMM_free(ConvexQP *problem, SolverOptions *options) → None[source]¶

siconos.numerics.convexQP_ADMM_init(ConvexQP *problem, SolverOptions *options) → None[source]¶

siconos.numerics.convexQP_ADMM_setDefaultSolverOptions(SolverOptions *options) → int[source]¶

set the default solver parameters and perform memory allocation for PG

Parameters:	options – the pointer to the array of options to set

siconos.numerics.convexQP_ProjectedGradient(ConvexQP *problem, array_like (np.float64, 1D)z, array_like (np.float64, 1D)w, int *info, SolverOptions *options) → None[source]¶

Projected Gradient solver for Convex QP problem.

Parameters:

problem – the variational inequality problem to solve z – global vector (n), in-out parameter w – global vector (n), in-out parameters info – return 0 if the solution is found options – the solver options : iparam[0] : Maximum iteration number dparam[3] (rho parameter. If rho >0, then self-adaptive (Armijo like)) – If rho <0, then constant rho parameter (rho <-- -rho) Adaptive step- (procedure.) – parameters (size) – 3.0/2.0; tauinv dparam[6] = 0.9; L dparam[7] = 0.3; Lmin (=) –

siconos.numerics.convexQP_ProjectedGradient_setDefaultSolverOptions(SolverOptions *options) → int[source]¶

set the default solver parameters and perform memory allocation for PG

Parameters:	options – the pointer to the array of options to set

siconos.numerics.convexQP_VI_solver(ConvexQP *problem, array_like (np.float64, 1D)z, array_like (np.float64, 1D)w, int *info, SolverOptions *options) → None[source]¶

siconos.numerics.convexQP_VI_solver_setDefaultSolverOptions(SolverOptions *options) → int[source]¶