Set the dual convergence tolerance of iterations.
Set the primal convergence tolerance of iterations.
set the regularization parameter.
set the regularization parameter. Default 0.0.
Set rho which is the augmented Lagrangian parameter.
Set rho which is the augmented Lagrangian parameter. kappa = regParam / (rho * numSubModels), if the absolute value of element in model is less than kappa, this element will be assigned to zero. So kappa should be less than 0.01 or 0.001.