iw.diaconis_fill¶
cal_beta_gamma¶
-
iw.diaconis_fill.cal_beta_gamma.
cal_beta_gamma
()¶ cal_beta_gamma function Computation of the estimation of gamma and beta
Inputs:
- Parameters
L (1d double array) – L is the laplacien matrix L nxn matrix; Markov generator
row (1d int_ array) – row array of sparse matrix L
col (1d int_ array) – column array of sparse matrix L
shape (int) – shape of Laplacien matrix
GXbarrebr – matrix (-L_Xbreve,Xbreve)^{-1}
row2 (1d int_ array) – row array of sparse matrix GXbarrebr
col2 (1d int_ array) – column array of sparse matrix GXbarrebr
shape2 (int) – shape of L matrix
Xbarre (1d int_ array) – vector of nR indices corresponding to the part of matrix L
Xbreve (1d int_ array) – vector of nR-n indices corresponding to complement of Xbarre the root indices
a (double) – maximum rate (maximum of the absolute value of the diagonal coefficients of L)
Output:
- Returns
tuple of (gam, beta) where gamma: numeric. 1/gamma= maximum Hitting time and beta: the mean of the return time after the first step.
- Return type
tuple of 2 double values
complementschur¶
-
iw.diaconis_fill.complementschur.
complementschur
()¶ Schur complement is the computation function M of D based on the formula: with
the Schur complement of D is
Input:
- Parameters
L (1d double array) – L is the laplacien matrix
row (1d int_ array) – row array of sparse matrix L
col (1d int_ array) – column array of sparse matrix L
shape (int) – shape of L matrix
R (1d int_ array) – vector of nR indices corresponding to the roots indices
Rc (1d int_ array) – vector of nR-n indices corresponding to complement of the root indices
Output:
- Returns
tuple of (Lbarre1d, row1, col1, shape1, abarre, GXbarrebr1d, row2, col2, shape2) where Lbarre1d is Schur complement of [L]_Rc in L , row1, col1, shape1, corresponding rox, col and shape abarre value of abarre, GXbarrebr1d ([L]_Rc)^{-1} 1d sparse matrix , row2, col2, shape2, corresponding rox, col and shape
- Return type
tuple of arrays and double values