def SpilsPrecSolve(self, t, u, fu, r, z, gamma, delta, lr, tmp):
"""
Preconditioner solve routine
"""
# Extract the P and pivot arrays from user_data.
P = self.P
pivot = self.pivot
zdata = z.data
r.Scale(ONE, z)
# Solve the block-diagonal system Px = r using LU factors stored
# in P and pivot data in pivot, and return the solution in z.
for jy in range(MY):
for jx in range(MX):
#v =
#print jx, jy, P[jx,jy,...], pivot[jx,jy,...], zdata[jx,jy,:]
denseGETRS(P[jx, jy, ...], pivot[jx, jy, ...], zdata[jx,
jy, :])
#print jx, jy, P[jx,jy,...], pivot[jx,jy,...], zdata[jx,jy,:]
return 0
def SpilsPrecSolve(self, t, u, fu, r, z, gamma, delta, lr, tmp):
"""
Preconditioner solve routine
"""
# Extract the P and pivot arrays from user_data.
P = self.P
pivot = self.pivot
zdata = z.data
r.Scale(ONE, z)
# Solve the block-diagonal system Px = r using LU factors stored
# in P and pivot data in pivot, and return the solution in z.
for jy in range(MY):
for jx in range(MX):
#v =
#print jx, jy, P[jx,jy,...], pivot[jx,jy,...], zdata[jx,jy,:]
denseGETRS(P[jx,jy,...], pivot[jx,jy,...], zdata[jx,jy,:])
#print jx, jy, P[jx,jy,...], pivot[jx,jy,...], zdata[jx,jy,:]
return 0
def testDenseGET():
"""
Simple test of denseGETRF and denseGETRS to solve a linear system
"""
N = 2
A = np.array([[4.0, 6.0], [3.0, 3.0]], dtype=np.float64, order='F')
A_sp = np.ascontiguousarray(A)
P = np.arange(N, dtype=np.int32)
b = np.ones(N, dtype=np.float64)
b_sp = linalg.lu_solve(linalg.lu_factor(A_sp), b)
ret = denseGETRF(A, P)
assert ret == 0
denseGETRS(A, P, b)
assert_allclose(b, b_sp)
assert ret == 0
def testDenseGET():
"""
Simple test of denseGETRF and denseGETRS to solve a linear system
"""
N = 2
A = np.array( [[4.0,6.0],[3.0,3.0]], dtype=np.float64, order='F')
A_sp = np.ascontiguousarray(A)
P = np.arange( N, dtype=np.int32 )
b = np.ones( N, dtype=np.float64 )
b_sp = linalg.lu_solve(linalg.lu_factor(A_sp), b)
ret = denseGETRF(A, P)
assert ret == 0
denseGETRS(A, P, b)
assert_allclose(b, b_sp)
assert ret == 0
def SpilsPrecSolve(self, tn, c, fc, r, z, gamma, delta, lr, vtemp):
"""
This routine applies two inverse preconditioner matrices
to the vector r, using the interaction-only block-diagonal Jacobian
with block-grouping, denoted Jr, and Gauss-Seidel applied to the
diffusion contribution to the Jacobian, denoted Jd.
It first calls GSIter for a Gauss-Seidel approximation to
((I - gamma*Jd)-inverse)*r, and stores the result in z.
Then it computes ((I - gamma*Jr)-inverse)*z, using LU factors of the
blocks in P, and pivot information in pivot, and returns the result in z.
"""
r.Scale(ONE, z)
# call GSIter for Gauss-Seidel iterations
self.GSIter(gamma, z, vtemp)
# Do backsolves for inverse of block-diagonal preconditioner factor
P = self.P
pivot = self.pivot
mx = self.mx
my = self.my
#ngx = self.ngx
mp = self.mp
jigx = self.jigx
jigy = self.jigy
iv = 0
for jy in range(my):
igy = jigy[jy]
for jx in range(mx):
igx = jigx[jx]
#ig = igx + igy*ngx
denseGETRS(P[igx, igy, ...], mp, pivot[igx, igy, :],
z.data[iv])
iv += mp
return 0
