sggsvd.f.html

来自「famous linear algebra library (LAPACK) p」· HTML 代码 · 共 360 行 · 第 1/2 页

HTML
360
字号
<!DOCTYPE HTML PUBLIC "-//W3C//DTD HTML 4.01 Transitional//EN">
<html>
 <head>
  <title>sggsvd.f</title>
 <meta name="generator" content="emacs 21.3.1; htmlfontify 0.20">
<style type="text/css"><!-- 
body { background: rgb(255, 255, 255);  color: rgb(0, 0, 0);  font-style: normal;  font-weight: 500;  font-stretch: normal;  font-family: adobe-courier;  font-size: 11pt;  text-decoration: none; }
span.default   { background: rgb(255, 255, 255);  color: rgb(0, 0, 0);  font-style: normal;  font-weight: 500;  font-stretch: normal;  font-family: adobe-courier;  font-size: 11pt;  text-decoration: none; }
span.default a { background: rgb(255, 255, 255);  color: rgb(0, 0, 0);  font-style: normal;  font-weight: 500;  font-stretch: normal;  font-family: adobe-courier;  font-size: 11pt;  text-decoration: underline; }
span.string   { color: rgb(188, 143, 143);  background: rgb(255, 255, 255);  font-style: normal;  font-weight: 500;  font-stretch: normal;  font-family: adobe-courier;  font-size: 11pt;  text-decoration: none; }
span.string a { color: rgb(188, 143, 143);  background: rgb(255, 255, 255);  font-style: normal;  font-weight: 500;  font-stretch: normal;  font-family: adobe-courier;  font-size: 11pt;  text-decoration: underline; }
span.comment   { color: rgb(178, 34, 34);  background: rgb(255, 255, 255);  font-style: normal;  font-weight: 500;  font-stretch: normal;  font-family: adobe-courier;  font-size: 11pt;  text-decoration: none; }
span.comment a { color: rgb(178, 34, 34);  background: rgb(255, 255, 255);  font-style: normal;  font-weight: 500;  font-stretch: normal;  font-family: adobe-courier;  font-size: 11pt;  text-decoration: underline; }
 --></style>

 </head>
  <body>

<pre>
      SUBROUTINE <a name="SGGSVD.1"></a><a href="sggsvd.f.html#SGGSVD.1">SGGSVD</a>( JOBU, JOBV, JOBQ, M, N, P, K, L, A, LDA, B,
     $                   LDB, ALPHA, BETA, U, LDU, V, LDV, Q, LDQ, WORK,
     $                   IWORK, INFO )
<span class="comment">*</span><span class="comment">
</span><span class="comment">*</span><span class="comment">  -- LAPACK driver routine (version 3.1) --
</span><span class="comment">*</span><span class="comment">     Univ. of Tennessee, Univ. of California Berkeley and NAG Ltd..
</span><span class="comment">*</span><span class="comment">     November 2006
</span><span class="comment">*</span><span class="comment">
</span><span class="comment">*</span><span class="comment">     .. Scalar Arguments ..
</span>      CHARACTER          JOBQ, JOBU, JOBV
      INTEGER            INFO, K, L, LDA, LDB, LDQ, LDU, LDV, M, N, P
<span class="comment">*</span><span class="comment">     ..
</span><span class="comment">*</span><span class="comment">     .. Array Arguments ..
</span>      INTEGER            IWORK( * )
      REAL               A( LDA, * ), ALPHA( * ), B( LDB, * ),
     $                   BETA( * ), Q( LDQ, * ), U( LDU, * ),
     $                   V( LDV, * ), WORK( * )
<span class="comment">*</span><span class="comment">     ..
</span><span class="comment">*</span><span class="comment">
</span><span class="comment">*</span><span class="comment">  Purpose
</span><span class="comment">*</span><span class="comment">  =======
</span><span class="comment">*</span><span class="comment">
</span><span class="comment">*</span><span class="comment">  <a name="SGGSVD.23"></a><a href="sggsvd.f.html#SGGSVD.1">SGGSVD</a> computes the generalized singular value decomposition (GSVD)
</span><span class="comment">*</span><span class="comment">  of an M-by-N real matrix A and P-by-N real matrix B:
</span><span class="comment">*</span><span class="comment">
</span><span class="comment">*</span><span class="comment">      U'*A*Q = D1*( 0 R ),    V'*B*Q = D2*( 0 R )
</span><span class="comment">*</span><span class="comment">
</span><span class="comment">*</span><span class="comment">  where U, V and Q are orthogonal matrices, and Z' is the transpose
</span><span class="comment">*</span><span class="comment">  of Z.  Let K+L = the effective numerical rank of the matrix (A',B')',
</span><span class="comment">*</span><span class="comment">  then R is a K+L-by-K+L nonsingular upper triangular matrix, D1 and
</span><span class="comment">*</span><span class="comment">  D2 are M-by-(K+L) and P-by-(K+L) &quot;diagonal&quot; matrices and of the
</span><span class="comment">*</span><span class="comment">  following structures, respectively:
</span><span class="comment">*</span><span class="comment">
</span><span class="comment">*</span><span class="comment">  If M-K-L &gt;= 0,
</span><span class="comment">*</span><span class="comment">
</span><span class="comment">*</span><span class="comment">                      K  L
</span><span class="comment">*</span><span class="comment">         D1 =     K ( I  0 )
</span><span class="comment">*</span><span class="comment">                  L ( 0  C )
</span><span class="comment">*</span><span class="comment">              M-K-L ( 0  0 )
</span><span class="comment">*</span><span class="comment">
</span><span class="comment">*</span><span class="comment">                    K  L
</span><span class="comment">*</span><span class="comment">         D2 =   L ( 0  S )
</span><span class="comment">*</span><span class="comment">              P-L ( 0  0 )
</span><span class="comment">*</span><span class="comment">
</span><span class="comment">*</span><span class="comment">                  N-K-L  K    L
</span><span class="comment">*</span><span class="comment">    ( 0 R ) = K (  0   R11  R12 )
</span><span class="comment">*</span><span class="comment">              L (  0    0   R22 )
</span><span class="comment">*</span><span class="comment">
</span><span class="comment">*</span><span class="comment">  where
</span><span class="comment">*</span><span class="comment">
</span><span class="comment">*</span><span class="comment">    C = diag( ALPHA(K+1), ... , ALPHA(K+L) ),
</span><span class="comment">*</span><span class="comment">    S = diag( BETA(K+1),  ... , BETA(K+L) ),
</span><span class="comment">*</span><span class="comment">    C**2 + S**2 = I.
</span><span class="comment">*</span><span class="comment">
</span><span class="comment">*</span><span class="comment">    R is stored in A(1:K+L,N-K-L+1:N) on exit.
</span><span class="comment">*</span><span class="comment">
</span><span class="comment">*</span><span class="comment">  If M-K-L &lt; 0,
</span><span class="comment">*</span><span class="comment">
</span><span class="comment">*</span><span class="comment">                    K M-K K+L-M
</span><span class="comment">*</span><span class="comment">         D1 =   K ( I  0    0   )
</span><span class="comment">*</span><span class="comment">              M-K ( 0  C    0   )
</span><span class="comment">*</span><span class="comment">
</span><span class="comment">*</span><span class="comment">                      K M-K K+L-M
</span><span class="comment">*</span><span class="comment">         D2 =   M-K ( 0  S    0  )
</span><span class="comment">*</span><span class="comment">              K+L-M ( 0  0    I  )
</span><span class="comment">*</span><span class="comment">                P-L ( 0  0    0  )
</span><span class="comment">*</span><span class="comment">
</span><span class="comment">*</span><span class="comment">                     N-K-L  K   M-K  K+L-M
</span><span class="comment">*</span><span class="comment">    ( 0 R ) =     K ( 0    R11  R12  R13  )
</span><span class="comment">*</span><span class="comment">                M-K ( 0     0   R22  R23  )
</span><span class="comment">*</span><span class="comment">              K+L-M ( 0     0    0   R33  )
</span><span class="comment">*</span><span class="comment">
</span><span class="comment">*</span><span class="comment">  where
</span><span class="comment">*</span><span class="comment">
</span><span class="comment">*</span><span class="comment">    C = diag( ALPHA(K+1), ... , ALPHA(M) ),
</span><span class="comment">*</span><span class="comment">    S = diag( BETA(K+1),  ... , BETA(M) ),
</span><span class="comment">*</span><span class="comment">    C**2 + S**2 = I.
</span><span class="comment">*</span><span class="comment">
</span><span class="comment">*</span><span class="comment">    (R11 R12 R13 ) is stored in A(1:M, N-K-L+1:N), and R33 is stored
</span><span class="comment">*</span><span class="comment">    ( 0  R22 R23 )
</span><span class="comment">*</span><span class="comment">    in B(M-K+1:L,N+M-K-L+1:N) on exit.
</span><span class="comment">*</span><span class="comment">
</span><span class="comment">*</span><span class="comment">  The routine computes C, S, R, and optionally the orthogonal
</span><span class="comment">*</span><span class="comment">  transformation matrices U, V and Q.
</span><span class="comment">*</span><span class="comment">
</span><span class="comment">*</span><span class="comment">  In particular, if B is an N-by-N nonsingular matrix, then the GSVD of
</span><span class="comment">*</span><span class="comment">  A and B implicitly gives the SVD of A*inv(B):
</span><span class="comment">*</span><span class="comment">                       A*inv(B) = U*(D1*inv(D2))*V'.
</span><span class="comment">*</span><span class="comment">  If ( A',B')' has orthonormal columns, then the GSVD of A and B is
</span><span class="comment">*</span><span class="comment">  also equal to the CS decomposition of A and B. Furthermore, the GSVD
</span><span class="comment">*</span><span class="comment">  can be used to derive the solution of the eigenvalue problem:
</span><span class="comment">*</span><span class="comment">                       A'*A x = lambda* B'*B x.
</span><span class="comment">*</span><span class="comment">  In some literature, the GSVD of A and B is presented in the form
</span><span class="comment">*</span><span class="comment">                   U'*A*X = ( 0 D1 ),   V'*B*X = ( 0 D2 )
</span><span class="comment">*</span><span class="comment">  where U and V are orthogonal and X is nonsingular, D1 and D2 are
</span><span class="comment">*</span><span class="comment">  ``diagonal''.  The former GSVD form can be converted to the latter
</span><span class="comment">*</span><span class="comment">  form by taking the nonsingular matrix X as
</span><span class="comment">*</span><span class="comment">
</span><span class="comment">*</span><span class="comment">                       X = Q*( I   0    )
</span><span class="comment">*</span><span class="comment">                             ( 0 inv(R) ).
</span><span class="comment">*</span><span class="comment">
</span><span class="comment">*</span><span class="comment">  Arguments
</span><span class="comment">*</span><span class="comment">  =========
</span><span class="comment">*</span><span class="comment">
</span><span class="comment">*</span><span class="comment">  JOBU    (input) CHARACTER*1
</span><span class="comment">*</span><span class="comment">          = 'U':  Orthogonal matrix U is computed;
</span><span class="comment">*</span><span class="comment">          = 'N':  U is not computed.
</span><span class="comment">*</span><span class="comment">
</span><span class="comment">*</span><span class="comment">  JOBV    (input) CHARACTER*1
</span><span class="comment">*</span><span class="comment">          = 'V':  Orthogonal matrix V is computed;
</span><span class="comment">*</span><span class="comment">          = 'N':  V is not computed.
</span><span class="comment">*</span><span class="comment">
</span><span class="comment">*</span><span class="comment">  JOBQ    (input) CHARACTER*1
</span><span class="comment">*</span><span class="comment">          = 'Q':  Orthogonal matrix Q is computed;
</span><span class="comment">*</span><span class="comment">          = 'N':  Q is not computed.
</span><span class="comment">*</span><span class="comment">
</span><span class="comment">*</span><span class="comment">  M       (input) INTEGER
</span><span class="comment">*</span><span class="comment">          The number of rows of the matrix A.  M &gt;= 0.
</span><span class="comment">*</span><span class="comment">
</span><span class="comment">*</span><span class="comment">  N       (input) INTEGER
</span><span class="comment">*</span><span class="comment">          The number of columns of the matrices A and B.  N &gt;= 0.
</span><span class="comment">*</span><span class="comment">
</span><span class="comment">*</span><span class="comment">  P       (input) INTEGER
</span><span class="comment">*</span><span class="comment">          The number of rows of the matrix B.  P &gt;= 0.
</span><span class="comment">*</span><span class="comment">
</span><span class="comment">*</span><span class="comment">  K       (output) INTEGER
</span><span class="comment">*</span><span class="comment">  L       (output) INTEGER
</span><span class="comment">*</span><span class="comment">          On exit, K and L specify the dimension of the subblocks
</span><span class="comment">*</span><span class="comment">          described in the Purpose section.
</span><span class="comment">*</span><span class="comment">          K + L = effective numerical rank of (A',B')'.
</span><span class="comment">*</span><span class="comment">
</span><span class="comment">*</span><span class="comment">  A       (input/output) REAL array, dimension (LDA,N)
</span><span class="comment">*</span><span class="comment">          On entry, the M-by-N matrix A.
</span><span class="comment">*</span><span class="comment">          On exit, A contains the triangular matrix R, or part of R.
</span><span class="comment">*</span><span class="comment">          See Purpose for details.
</span><span class="comment">*</span><span class="comment">
</span><span class="comment">*</span><span class="comment">  LDA     (input) INTEGER
</span><span class="comment">*</span><span class="comment">          The leading dimension of the array A. LDA &gt;= max(1,M).
</span><span class="comment">*</span><span class="comment">
</span><span class="comment">*</span><span class="comment">  B       (input/output) REAL array, dimension (LDB,N)
</span><span class="comment">*</span><span class="comment">          On entry, the P-by-N matrix B.
</span><span class="comment">*</span><span class="comment">          On exit, B contains the triangular matrix R if M-K-L &lt; 0.
</span><span class="comment">*</span><span class="comment">          See Purpose for details.
</span><span class="comment">*</span><span class="comment">
</span><span class="comment">*</span><span class="comment">  LDB     (input) INTEGER
</span><span class="comment">*</span><span class="comment">          The leading dimension of the array B. LDB &gt;= max(1,P).
</span><span class="comment">*</span><span class="comment">
</span><span class="comment">*</span><span class="comment">  ALPHA   (output) REAL array, dimension (N)
</span><span class="comment">*</span><span class="comment">  BETA    (output) REAL array, dimension (N)
</span><span class="comment">*</span><span class="comment">          On exit, ALPHA and BETA contain the generalized singular
</span><span class="comment">*</span><span class="comment">          value pairs of A and B;
</span><span class="comment">*</span><span class="comment">            ALPHA(1:K) = 1,
</span><span class="comment">*</span><span class="comment">            BETA(1:K)  = 0,
</span><span class="comment">*</span><span class="comment">          and if M-K-L &gt;= 0,
</span><span class="comment">*</span><span class="comment">            ALPHA(K+1:K+L) = C,
</span><span class="comment">*</span><span class="comment">            BETA(K+1:K+L)  = S,
</span><span class="comment">*</span><span class="comment">          or if M-K-L &lt; 0,
</span><span class="comment">*</span><span class="comment">            ALPHA(K+1:M)=C, ALPHA(M+1:K+L)=0
</span><span class="comment">*</span><span class="comment">            BETA(K+1:M) =S, BETA(M+1:K+L) =1
</span><span class="comment">*</span><span class="comment">          and
</span><span class="comment">*</span><span class="comment">            ALPHA(K+L+1:N) = 0

⌨️ 快捷键说明

复制代码Ctrl + C
搜索代码Ctrl + F
全屏模式F11
增大字号Ctrl + =
减小字号Ctrl + -
显示快捷键?