f02wgc returns leading terms in the singular value decomposition (SVD) of a real general matrix and computes the corresponding left and right singular vectors.
The function may be called by the names: f02wgc, nag_eigen_real_gen_partialsvd or nag_real_partial_svd.
3Description
f02wgc computes a few, , of the largest singular values and corresponding vectors of an matrix . The value of should be small relative to and , for example . The full singular value decomposition (SVD) of an matrix is given by
where and are orthogonal and is an diagonal matrix with real diagonal elements, , such that
The are the singular values of and the first columns of and are the left and right singular vectors of .
If , denote the leading columns of and respectively, and if denotes the leading principal submatrix of , then
is the best rank- approximation to in both the -norm and the Frobenius norm.
The singular values and singular vectors satisfy
where and are the th columns of and respectively.
Thus, for , the largest singular values and corresponding right singular vectors are computed by finding eigenvalues and eigenvectors for the symmetric matrix . For , the largest singular values and corresponding left singular vectors are computed by finding eigenvalues and eigenvectors for the symmetric matrix . These eigenvalues and eigenvectors are found using functions from Chapter F12. You should read the F12 Chapter Introduction for full details of the method used here.
The real matrix is not explicitly supplied to f02wgc. Instead, you are required to supply a function, av, that must calculate one of the requested matrix-vector products or for a given real vector (of length or respectively).
4References
Wilkinson J H (1978) Singular Value Decomposition – Basic Aspects Numerical Software – Needs and Availability (ed D A H Jacobs) Academic Press
5Arguments
1: – Nag_OrderTypeInput
On entry: the order argument specifies the two-dimensional storage scheme being used, i.e., row-major ordering or column-major ordering. C language defined storage is specified by . See Section 3.1.3 in the Introduction to the NAG Library CL Interface for a more detailed explanation of the use of this argument.
Constraint:
or .
2: – IntegerInput
On entry: , the number of rows of the matrix .
Constraint:
.
If , an immediate return is effected.
3: – IntegerInput
On entry: , the number of columns of the matrix .
Constraint:
.
If , an immediate return is effected.
4: – IntegerInput
On entry: , the number of singular values to be computed.
Constraint:
.
5: – IntegerInput
On entry: the dimension of the arrays sigma and resid.
This is the number of Lanczos basis vectors to use during the computation of the largest eigenvalues of () or ().
At present there is no a priori analysis to guide the selection of ncv relative to k. However, it is recommended that . If many problems of the same type are to be solved, you should experiment with varying ncv while keeping k fixed for a given test problem. This will usually decrease the required number of matrix-vector operations but it also increases the internal storage required to maintain the orthogonal basis vectors. The optimal ‘cross-over’ with respect to CPU time is problem dependent and must be determined empirically.
Constraint:
.
6: – function, supplied by the userExternal Function
av must return the vector result of the matrix-vector product or , as indicated by the input value of iflag, for the given vector .
av (Integer *iflag,Integer m,Integer n,const double x[],double ax[],Nag_Comm *comm)
1: – Integer *Input/Output
On entry: if , ax must return the -vector result of the matrix-vector product .
If , ax must return the -vector result of the matrix-vector product .
On exit: may be used as a flag to indicate a failure in the computation of or . If iflag is negative on exit from av, f02wgc will exit immediately with fail set to iflag.
On exit: if , contains the -vector result of the matrix-vector product .
If , contains the -vector result of the matrix-vector product .
6: – Nag_Comm *
Pointer to structure of type Nag_Comm; the following members are relevant to av.
user – double *
iuser – Integer *
p – Pointer
The type Pointer will be void *. Before calling f02wgc you may allocate memory and initialize these pointers with various quantities for use by av when called from f02wgc (see Section 3.1.1 in the Introduction to the NAG Library CL Interface).
Note:av should not return floating-point NaN (Not a Number) or infinity values, since these are not handled by f02wgc. If your code inadvertently does return any NaNs or infinities, f02wgc is likely to produce unexpected results.
7: – Integer *Output
On exit: the number of converged singular values found.
8: – doubleOutput
On exit: the nconv converged singular values are stored in the first nconv elements of sigma.
9: – doubleOutput
Note: the dimension, dim, of the array
u
must be at least
when ;
when .
where appears in this document, it refers to the array element
when ;
when .
On exit: the left singular vectors corresponding to the singular values stored in sigma.
The
th element of the th left singular vector is stored in , for and .
On exit: the residual , for , or , for , for each of the converged singular values and corresponding left and right singular vectors and .
14: – Nag_Comm *
The NAG communication argument (see Section 3.1.1 in the Introduction to the NAG Library CL Interface).
15: – NagError *Input/Output
The NAG error argument (see Section 7 in the Introduction to the NAG Library CL Interface).
f02wgc returns with NE_NOERROR if at least singular values have converged and the corresponding left and right singular vectors have been computed.
6Error Indicators and Warnings
NE_ALLOC_FAIL
Dynamic memory allocation failed.
See Section 3.1.2 in the Introduction to the NAG Library CL Interface for further information.
NE_BAD_PARAM
On entry, argument had an illegal value.
NE_EIGENVALUES
The number of eigenvalues found to sufficient accuracy is zero.
NE_INT
On entry, .
Constraint: .
On entry, .
Constraint: .
On entry, .
Constraint: .
On entry, . Constraint: .
On entry, . Constraint: .
NE_INT_2
On entry, and .
Constraint: .
On entry, and .
Constraint: .
On entry, and .
Constraint: .
On entry, and .
Constraint: .
NE_INT_3
On entry, , and .
Constraint: .
NE_INT_4
On entry, , , and .
Constraint: .
NE_INTERNAL_ERROR
An error occurred during an internal call. Consider increasing the size of ncv relative to k.
An internal error has occurred in this function. Check the function call and any array sizes. If the call is correct then please contact NAG for assistance.
See Section 7.5 in the Introduction to the NAG Library CL Interface for further information.
NE_LANCZOS_ITERATION
No shifts could be applied during a cycle of the implicitly restarted Lanczos iteration.
NE_MAX_ITER
The maximum number of iterations has been reached. The maximum number of iterations . The number of converged eigenvalues .
NE_NO_LANCZOS_FAC
Could not build a full Lanczos factorization.
NE_NO_LICENCE
Your licence key may have expired or may not have been installed correctly.
See Section 8 in the Introduction to the NAG Library CL Interface for further information.
NE_USER_STOP
On output from user-defined function av, iflag was set to a negative value, .
Background information to multithreading can be found in the Multithreading documentation.
f02wgc is threaded by NAG for parallel execution in multithreaded implementations of the NAG Library.
f02wgc makes calls to BLAS and/or LAPACK routines, which may be threaded within the vendor library used by this implementation. Consult the documentation for the vendor library for further information.
Please consult the X06 Chapter Introduction for information on how to control and interrogate the OpenMP environment used within this function. Please also consult the Users' Note for your implementation for any additional implementation-specific information.
9Further Comments
None.
10Example
This example finds the four largest singular values () and corresponding right and left singular vectors for the matrix , where is the real matrix derived from the simplest finite difference discretization of the two-dimensional kernel where