Dew MtxVec NET
TSparseMtx Object
Summary
Encapsulates sparse matrix properties and methods/operations.

Unit
Sparse

Hierarchy
TMtxVec <--TSparseMtx

Subclasses
None

Description
The class encapsulates sparse matrix properties and methods/operations like sparse matrix addition, subtraction, multiplication, mixed operations with dense vectors and matrices, three types of sparse linear system solvers: Umfpack, Pardiso and Iterative, different matrix storage format conversion methods and more...

Properties

 Name  Summary 
ActualIters Returns the number of actual iterations performed by Solve
AutoClearReport Defines if call to Solve will clear the report. 
AutoPattern  
BackError Holds the backward or iterative stopping criteria. 
Cols Number of columns in the sparse matrix. 
ConvergenceCheck Convergence criteria for iterative method. 
CValue  
DontStripZeros Prevents zeros to be stripped. 
IterativeMethod Defines the iterative method for solving system of equations. 
MaxIters Upper limit for number of iterations. 
MtxError Holds the error string, if an error has occurred. 
NonZeros Defines the number of non-zero elements in the sparse matrix. 
Pardiso Pardiso settings. 
RefinmentIters Defines number of refinement steps the Solve method should try. 
ReportLevel Defines how detailed report will be printed. 
Rows Number of rows in the sparse matrix. 
SparseBlockSize The size of the block as used by UmfPack. 
SparsePattern Defines the pattern used when solving a system of linear equations. 
SparseSolver Specifies the sparse matrix solver to use. 
SparseSystem Defines the system to be solved by the Solve method. 
Taucs Taucs settings. 
Tolerance Defines the tolerance for iterative method. 
Value Allows setting and getting value at position Row and Col. 
VectorCount Defines maximum number of vectors used in some iterative methods. 

Methods

 Name  Summary 
Add Sums two sparse matrices (A+B). 
AddSplit Construct sparse matrxix by summing lower, upper triangluar part and the diagonal. 
Adjung Adjungate operation. 
Assign  
BandedToSparse Converts banded matrix to modified compressed column sparse matrix format. 
CachePointer  
ConjMul Conjugate Sparse2 and matrix multiply it with Sparse1. 
Copy Copies sparse matrix from Src to the calling object. 
DenseToSparse Convert dense matrix to sparse matrix format. 
Diag Sets the kk-th diagonal of TSparseMtx to the Vec object. 
DiagonalsCount Calculates the number of diagonals above and below main diagonal. 
DivideElem Divide individual elements of sparse matrices. 
dmpl  
DoSetCols  
DoSetComplex  
Equal Compares two sparse matrices. 
ExpandSymmetric Form a full symmetric matrix from upper half with diagonal included. 
Finalize  
FreeNumeric Frees the numeric object used by the Umfpack. 
InternalFreeSymbolic  
Inv The inverse matrix. 
IsEqualSize Compares matrix size. 
LoadFromHarwellBoeingFile  
LoadFromMatrixMarketFile Loads the matrix stored in a matrix market file format. 
LowerTriangle Extract lower part of the Mtx. 
Max Maximum of all non-zero elements. 
Maxc Maximum of all non-zero complex elements. 
Mean Average of all non-zero elements. 
Meanc Average of all non-zero complex elements. 
Min Minimum of all non-zero elements. 
Minc Minimum of all non-zero complex elements. 
MtxTypePardiso  
MtxTypeTaucs  
Mul Multiply two sparse matrices (matrix multiplication). 
MulLeft Multiply the sparse matrix from left. 
MulRight Multiply the sparse matrix from right. 
PixelDownSample Reduces dense matrix for screen display (bitmap) to show the non-zero pattern. 
RandomSparse Creates quadratic non-complex non-singular sparse matrix. 
ResidualNorm The residual norm of the solution of the system of linear equations A*X=B. 
Resize Resizes the sparse matrix to a new number of non-zeros. 
RMS Root mean squared of all non-zero elements. 
SaveToHarwellBoeingFile  
SaveToMatrixMarketFile Saves the matrix in matrix market file format. 
setActualSize  
SetLength  
SetPardiso  
SetTaucs  
Size Define the size of sparse matrix. 
Solve Vector version of Solve. 
SolveIterative Solve the system A*X = B by using one of the iterative methods. 
SolvePardiso  
SolvePardisoInternal  
SolveTaucs  
SolveTaucsInternal  
SolveUmfPack  
SparseToBanded Convert sparse matrix to banded matrix. 
SparseToDense Convert sparse matrix to banded matrix. 
SparseToTriplets Convert sparse matrix to triplets. 
Split Split sparse matrix into lower triangle, main diagonal and upper triangle. 
StringsToValues Convert strings to specific matrix format. 
StructurizeSymmetric Required by Pardiso solver to solve for general matrices. 
Sub Subract two sparse matrices. 
Transp Transpose sparse matrix. 
TripletsToSparse Convert triplets to sparse format. 
TripletsToSparseAndFlush Construct a sparse matrix by specifying the position of each element in dense matrix. 
UpperTriangle Extract upper part of the Mtx. 
ValuesToStrings Convert matrix values to strings. 
WriteHeader  
ZeroThresh Reduces the number of non-zeros. 

Copyright 2008 Dew Research
http://www.dewresearch.com