Class Pfqn_egflinearizer

java.lang.Object
jline.api.pfqn.mva.Pfqn_egflinearizer

public final class Pfqn_egflinearizer extends Object
  • Method Details

    • pfqn_egflinearizer

      public static Ret.pfqnAMVA pfqn_egflinearizer(Matrix L, Matrix N, Matrix Z, SchedStrategy[] type, double tol, int maxiter, Matrix alpha)
    • pfqn_egflinearizer

      public static Ret.pfqnAMVA pfqn_egflinearizer(Matrix L, Matrix N, Matrix Z, SchedStrategy[] type, double tol, int maxiter, Matrix alpha, Matrix QN0)
    • pfqn_egflinearizer

      public static Ret.pfqnAMVA pfqn_egflinearizer(Matrix L, Matrix N, Matrix Z, SchedStrategy[] type, double tol, int maxiter, Matrix alpha, Matrix QN0, int npasses)
      Parameters:
      tol - convergence tolerance on the Frobenius norm of dQ; NaN selects the published Linearizer termination test of Chandy and Neuse, Commun. ACM 25(2), 1982, p.129, under which each Core call stops when max_{i,r}|dQ(i,r)|/N_r falls below Pfqn_cntol.pfqn_cntol evaluated at the population Core is running at
      npasses - number of Delta refresh passes; 3 is the Chandy-Neuse fixed rule, Pfqn_scat passes 1