菜鸟求助：IVF中实矩阵求逆用哪个函数啊？

sailor8

vvt 发表于 2017-1-2 08:12
1. 用 matmul 代替 .x. 进行矩阵相乘。
2. 用 IMSL 5.0 代替 6.0和7.0（因为它们有bug）
3. 用 MKL 、Lapac ...

好，我尝试一下imsl5.0。
谢谢了～

kerb

subroutine dgesv        (        integer         N,
integer         NRHS,
double precision, dimension( lda, * )         A,
integer         LDA,
integer, dimension( * )         IPIV,
double precision, dimension( ldb, * )         B,
integer         LDB,
integer         INFO
)               
DGESV computes the solution to system of linear equations A * X = B for GE matrices

Download DGESV + dependencies [TGZ] [ZIP] [TXT]

Purpose:
DGESV computes the solution to a real system of linear equations
    A * X = B,
where A is an N-by-N matrix and X and B are N-by-NRHS matrices.

The LU decomposition with partial pivoting and row interchanges is
used to factor A as
    A = P * L * U,
where P is a permutation matrix, L is unit lower triangular, and U is
upper triangular.  The factored form of A is then used to solve the
system of equations A * X = B.
Parameters
[in]        N       
          N is INTEGER
          The number of linear equations, i.e., the order of the
          matrix A.  N >= 0.
[in]        NRHS       
          NRHS is INTEGER
          The number of right hand sides, i.e., the number of columns
          of the matrix B.  NRHS >= 0.
[in,out]        A       
          A is DOUBLE PRECISION array, dimension (LDA,N)
          On entry, the N-by-N coefficient matrix A.
          On exit, the factors L and U from the factorization
          A = P*L*U; the unit diagonal elements of L are not stored.
[in]        LDA       
          LDA is INTEGER
          The leading dimension of the array A.  LDA >= max(1,N).
[out]        IPIV       
          IPIV is INTEGER array, dimension (N)
          The pivot indices that define the permutation matrix P;
          row i of the matrix was interchanged with row IPIV(i).
[in,out]        B       
          B is DOUBLE PRECISION array, dimension (LDB,NRHS)
          On entry, the N-by-NRHS matrix of right hand side matrix B.
          On exit, if INFO = 0, the N-by-NRHS solution matrix X.
[in]        LDB       
          LDB is INTEGER
          The leading dimension of the array B.  LDB >= max(1,N).
[out]        INFO       
          INFO is INTEGER
          = 0:  successful exit
          < 0:  if INFO = -i, the i-th argument had an illegal value
          > 0:  if INFO = i, U(i,i) is exactly zero.  The factorization
                has been completed, but the factor U is exactly
                singular, so the solution could not be computed.