ROOT 6.13/01 Reference Guide |
Single Value Decomposition class.
For an (m x n) matrix A with m >= n, the singular value decomposition is an (m x m) orthogonal matrix fU, an (m x n) diagonal matrix fS, and an (n x n) orthogonal matrix fV so that A = U*S*V'.
If the row/column index of A starts at (rowLwb,colLwb) then the decomposed matrices/vectors start at :
The diagonal matrix fS is stored in the singular values vector fSig . The singular values, fSig[k] = S[k][k], are ordered so that fSig[0] >= fSig[1] >= ... >= fSig[n-1].
The singular value decomposition always exists, so the decomposition will (as long as m >=n) never fail. If m < n, the user should add sufficient zero rows to A , so that m == n
Here fTol is used to set the threshold on the minimum allowed value of the singular values: min_singular = fTol*max(fSig[i])
Definition at line 23 of file TDecompSVD.h.
Public Types | |
enum | { kWorkMax = 100 } |
Public Member Functions | |
TDecompSVD () | |
TDecompSVD (Int_t nrows, Int_t ncols) | |
Constructor for (nrows x ncols) matrix. More... | |
TDecompSVD (Int_t row_lwb, Int_t row_upb, Int_t col_lwb, Int_t col_upb) | |
Constructor for ([row_lwb..row_upb] x [col_lwb..col_upb]) matrix. More... | |
TDecompSVD (const TMatrixD &m, Double_t tol=0.0) | |
Constructor for general matrix A . More... | |
TDecompSVD (const TDecompSVD &another) | |
Copy constructor. More... | |
virtual | ~TDecompSVD () |
virtual Double_t | Condition () |
Matrix condition number. More... | |
virtual Bool_t | Decompose () |
SVD decomposition of matrix If the decomposition succeeds, bit kDecomposed is set , otherwise kSingular. More... | |
virtual void | Det (Double_t &d1, Double_t &d2) |
Matrix determinant det = d1*TMath::Power(2.,d2) More... | |
const TMatrixD | GetMatrix () |
Reconstruct the original matrix using the decomposition parts. More... | |
virtual Int_t | GetNcols () const |
virtual Int_t | GetNrows () const |
const TVectorD & | GetSig () |
const TMatrixD & | GetU () |
const TMatrixD & | GetV () |
Bool_t | Invert (TMatrixD &inv) |
For a matrix A(m,n), its inverse A_inv is defined as A * A_inv = A_inv * A = unit The user should always supply a matrix of size (m x m) ! If m > n , only the (n x m) part of the returned (pseudo inverse) matrix should be used . More... | |
TMatrixD | Invert (Bool_t &status) |
For a matrix A(m,n), its inverse A_inv is defined as A * A_inv = A_inv * A = unit (n x m) Ainv is returned . More... | |
TMatrixD | Invert () |
TDecompSVD & | operator= (const TDecompSVD &source) |
Assignment operator. More... | |
void | Print (Option_t *opt="") const |
Print class members. More... | |
virtual void | SetMatrix (const TMatrixD &a) |
Set matrix to be decomposed. More... | |
virtual Bool_t | Solve (TVectorD &b) |
Solve Ax=b assuming the SVD form of A is stored . More... | |
virtual TVectorD | Solve (const TVectorD &b, Bool_t &ok) |
virtual Bool_t | Solve (TMatrixDColumn &b) |
Solve Ax=b assuming the SVD form of A is stored . More... | |
virtual Bool_t | TransSolve (TVectorD &b) |
Solve A^T x=b assuming the SVD form of A is stored . Solution returned in b. More... | |
virtual TVectorD | TransSolve (const TVectorD &b, Bool_t &ok) |
virtual Bool_t | TransSolve (TMatrixDColumn &b) |
Solve A^T x=b assuming the SVD form of A is stored . Solution returned in b. More... | |
Public Member Functions inherited from TDecompBase | |
TDecompBase () | |
Default constructor. More... | |
TDecompBase (const TDecompBase &another) | |
Copy constructor. More... | |
virtual | ~TDecompBase () |
Int_t | GetColLwb () const |
Double_t | GetCondition () const |
Double_t | GetDet1 () const |
Double_t | GetDet2 () const |
Int_t | GetRowLwb () const |
Double_t | GetTol () const |
virtual Bool_t | MultiSolve (TMatrixD &B) |
Solve set of equations with RHS in columns of B. More... | |
TDecompBase & | operator= (const TDecompBase &source) |
Assignment operator. More... | |
void | Print (Option_t *opt="") const |
Print class members. More... | |
Double_t | SetTol (Double_t tol) |
Protected Member Functions | |
virtual const TMatrixDBase & | GetDecompMatrix () const |
Protected Member Functions inherited from TDecompBase | |
Int_t | Hager (Double_t &est, Int_t iter=5) |
void | ResetStatus () |
Static Protected Member Functions | |
static Bool_t | Bidiagonalize (TMatrixD &v, TMatrixD &u, TVectorD &sDiag, TVectorD &oDiag) |
Bidiagonalize the (m x n) - matrix a (stored in v) through a series of Householder transformations applied to the left (Q^T) and to the right (H) of a , so that A = Q . More... | |
static void | Diag_1 (TMatrixD &v, TVectorD &sDiag, TVectorD &oDiag, Int_t k) |
Step 1 in the matrix diagonalization. More... | |
static void | Diag_2 (TVectorD &sDiag, TVectorD &oDiag, Int_t k, Int_t l) |
Step 2 in the matrix diagonalization. More... | |
static void | Diag_3 (TMatrixD &v, TMatrixD &u, TVectorD &sDiag, TVectorD &oDiag, Int_t k, Int_t l) |
Step 3 in the matrix diagonalization. More... | |
static Bool_t | Diagonalize (TMatrixD &v, TMatrixD &u, TVectorD &sDiag, TVectorD &oDiag) |
Diagonalizes in an iterative fashion the bidiagonal matrix C as described through sDiag and oDiag, so that S' = U'^T . More... | |
static void | SortSingular (TMatrixD &v, TMatrixD &u, TVectorD &sDiag) |
Perform a permutation transformation on the diagonal matrix S', so that matrix S'' = U''^T . More... | |
Static Protected Member Functions inherited from TDecompBase | |
static void | DiagProd (const TVectorD &diag, Double_t tol, Double_t &d1, Double_t &d2) |
Protected Attributes | |
TVectorD | fSig |
TMatrixD | fU |
TMatrixD | fV |
Protected Attributes inherited from TDecompBase | |
Int_t | fColLwb |
Double_t | fCondition |
Double_t | fDet1 |
Double_t | fDet2 |
Int_t | fRowLwb |
Double_t | fTol |
Additional Inherited Members | |
Protected Types inherited from TDecompBase | |
enum | { kWorkMax = 100 } |
enum | EMatrixDecompStat { kInit = BIT(14), kPatternSet = BIT(15), kValuesSet = BIT(16), kMatrixSet = BIT(17), kDecomposed = BIT(18), kDetermined = BIT(19), kCondition = BIT(20), kSingular = BIT(21) } |
#include <TDecompSVD.h>
anonymous enum |
Enumerator | |
---|---|
kWorkMax |
Definition at line 43 of file TDecompSVD.h.
|
inline |
Definition at line 45 of file TDecompSVD.h.
TDecompSVD::TDecompSVD | ( | Int_t | nrows, |
Int_t | ncols | ||
) |
Constructor for (nrows x ncols) matrix.
Definition at line 52 of file TDecompSVD.cxx.
TDecompSVD::TDecompSVD | ( | Int_t | row_lwb, |
Int_t | row_upb, | ||
Int_t | col_lwb, | ||
Int_t | col_upb | ||
) |
Constructor for ([row_lwb..row_upb] x [col_lwb..col_upb]) matrix.
Definition at line 66 of file TDecompSVD.cxx.
TDecompSVD::TDecompSVD | ( | const TMatrixD & | m, |
Double_t | tol = 0.0 |
||
) |
Constructor for general matrix A .
Definition at line 85 of file TDecompSVD.cxx.
TDecompSVD::TDecompSVD | ( | const TDecompSVD & | another | ) |
Copy constructor.
Definition at line 114 of file TDecompSVD.cxx.
|
inlinevirtual |
Definition at line 50 of file TDecompSVD.h.
|
staticprotected |
Bidiagonalize the (m x n) - matrix a (stored in v) through a series of Householder transformations applied to the left (Q^T) and to the right (H) of a , so that A = Q .
C . H^T with matrix C bidiagonal. Q and H are orthogonal matrices .
Output: v - (n x n) - matrix H in the (n x n) part of v u - (m x m) - matrix Q^T sDiag - diagonal of the (m x n) C oDiag - off-diagonal elements of matrix C
Test code for the output:
Definition at line 192 of file TDecompSVD.cxx.
|
virtual |
Matrix condition number.
Reimplemented from TDecompBase.
Definition at line 823 of file TDecompSVD.cxx.
|
virtual |
SVD decomposition of matrix If the decomposition succeeds, bit kDecomposed is set , otherwise kSingular.
Implements TDecompBase.
Definition at line 123 of file TDecompSVD.cxx.
|
virtual |
Matrix determinant det = d1*TMath::Power(2.,d2)
Reimplemented from TDecompBase.
Definition at line 846 of file TDecompSVD.cxx.
|
staticprotected |
Step 1 in the matrix diagonalization.
Definition at line 374 of file TDecompSVD.cxx.
Step 2 in the matrix diagonalization.
Definition at line 398 of file TDecompSVD.cxx.
|
staticprotected |
Step 3 in the matrix diagonalization.
Definition at line 416 of file TDecompSVD.cxx.
|
staticprotected |
Diagonalizes in an iterative fashion the bidiagonal matrix C as described through sDiag and oDiag, so that S' = U'^T .
C . V' is diagonal. U' and V' are orthogonal matrices .
Output:
sDiag - diagonal of the (m x n) S'
return convergence flag: 0 -> no convergence 1 -> convergence
Test code for the output:
Definition at line 307 of file TDecompSVD.cxx.
|
inlineprotectedvirtual |
Implements TDecompBase.
Definition at line 39 of file TDecompSVD.h.
const TMatrixD TDecompSVD::GetMatrix | ( | ) |
Reconstruct the original matrix using the decomposition parts.
Definition at line 557 of file TDecompSVD.cxx.
|
virtual |
Implements TDecompBase.
Definition at line 870 of file TDecompSVD.cxx.
|
virtual |
Implements TDecompBase.
Definition at line 865 of file TDecompSVD.cxx.
|
inline |
Definition at line 59 of file TDecompSVD.h.
|
inline |
Definition at line 55 of file TDecompSVD.h.
|
inline |
Definition at line 57 of file TDecompSVD.h.
Bool_t TDecompSVD::Invert | ( | TMatrixD & | inv | ) |
For a matrix A(m,n), its inverse A_inv is defined as A * A_inv = A_inv * A = unit The user should always supply a matrix of size (m x m) ! If m > n , only the (n x m) part of the returned (pseudo inverse) matrix should be used .
Definition at line 881 of file TDecompSVD.cxx.
TMatrixD TDecompSVD::Invert | ( | Bool_t & | status | ) |
For a matrix A(m,n), its inverse A_inv is defined as A * A_inv = A_inv * A = unit (n x m) Ainv is returned .
Definition at line 903 of file TDecompSVD.cxx.
|
inline |
Definition at line 82 of file TDecompSVD.h.
TDecompSVD & TDecompSVD::operator= | ( | const TDecompSVD & | source | ) |
Assignment operator.
Definition at line 930 of file TDecompSVD.cxx.
void TDecompSVD::Print | ( | Option_t * | opt = "" | ) | const |
Print class members.
Definition at line 919 of file TDecompSVD.cxx.
|
virtual |
Set matrix to be decomposed.
Definition at line 582 of file TDecompSVD.cxx.
|
virtual |
Solve Ax=b assuming the SVD form of A is stored .
Solution returned in b. If A is of size (m x n), input vector b should be of size (m), however, the solution, returned in b, will be in the first (n) elements .
For m > n , x is the least-squares solution of min(A . x - b)
Implements TDecompBase.
Definition at line 615 of file TDecompSVD.cxx.
Implements TDecompBase.
Definition at line 66 of file TDecompSVD.h.
|
virtual |
Solve Ax=b assuming the SVD form of A is stored .
Solution returned in the matrix column cb b. If A is of size (m x n), input vector b should be of size (m), however, the solution, returned in b, will be in the first (n) elements .
For m > n , x is the least-squares solution of min(A . x - b)
Implements TDecompBase.
Definition at line 670 of file TDecompSVD.cxx.
Perform a permutation transformation on the diagonal matrix S', so that matrix S'' = U''^T .
S' . V'' has diagonal elements ordered such that they do not increase.
Output:
Definition at line 497 of file TDecompSVD.cxx.
|
virtual |
Solve A^T x=b assuming the SVD form of A is stored . Solution returned in b.
Implements TDecompBase.
Definition at line 723 of file TDecompSVD.cxx.
Implements TDecompBase.
Definition at line 72 of file TDecompSVD.h.
|
virtual |
Solve A^T x=b assuming the SVD form of A is stored . Solution returned in b.
Implements TDecompBase.
Definition at line 772 of file TDecompSVD.cxx.
|
protected |
Definition at line 30 of file TDecompSVD.h.
|
protected |
Definition at line 28 of file TDecompSVD.h.
|
protected |
Definition at line 29 of file TDecompSVD.h.