1
2
3
4
5
6
7
8
9
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
42
43
44
45
46
47
48
49
|
#-------------------------------------------------------------------------------
# SPEX/Python/SPEXpy/cholesky_backslash.py: solve Ax=b using Cholesky factorization
#-------------------------------------------------------------------------------
# SPEX: (c) 2022-2024, Christopher Lourenco, Jinhao Chen,
# Lorena Mejia Domenzain, Erick Moreno-Centeno, and Timothy A. Davis.
# All Rights Reserved.
# SPDX-License-Identifier: GPL-2.0-or-later or LGPL-3.0-or-later
#------------------------------------------------------------------------------
from .Options import Options
from .SPEX_error import *
from .spex_connect import spex_connect
import scipy
from scipy.sparse import isspmatrix, isspmatrix_csc, linalg
def cholesky_backslash( A, b, options=Options('double', 'amd')):
## A is a scipy.sparse(data must be float64) #technically it only needs to be numerical
## b is a numpy.array (data must be float64)
## options is a dictionary that specifies what tipe the solution should be, this by default is double
##--------------------------------------------------------------------------
## Verify inputs
##--------------------------------------------------------------------------
if not isspmatrix(A):
raise SPEXerror(determine_error(3))
## If the sparse input matrix is not in csc form, convert it into csc form
if not isspmatrix_csc(A):
A.tocsc()
## Check symmetry
tol=1e-8
if linalg.norm(A-A.T, scipy.Inf) > tol:
raise SPEX_error(determine_error(-4))
# Check input shape
if A.shape[1]!=b.shape[0]:
raise SPEX_error(determine_error(-3))
if options.ordering==None:
options.default_chol()
##--------------------------------------------------------------------------
## Call SPEX
##--------------------------------------------------------------------------
x=spex_connect(A,b,options.order(),options.charOut(),4) #4 calls the LDL factorization
return x
|
