MGroup.LinearAlgebra.Commons

Conversions

Low level array operations to convert between matrix storage formats. Authors: Serafeim Bakalakos

public static class MGroup.LinearAlgebra.Commons.Conversions

Static Methods

Type	Name	Summary
Double[,]	Array2DLowerToSymmetric(Double[,] array2D)
Double[]	Array2DToFullColMajor(Double[,] array2D)
Double[]	Array2DToFullRowMajor(Double[,] array2D)
Double[]	Array2DToPackedLowerColMajor(Double[,] array2D)
Double[]	Array2DToPackedLowerRowMajor(Double[,] array2D)
Double[]	Array2DToPackedUpperColMajor(Double[,] array2D)
Double[]	Array2DToPackedUpperRowMajor(Double[,] array2D)
Double[,]	Array2DUpperToSymmetric(Double[,] array2D)
Double[]	ColumnMajorToRowMajor(Double[] colMajor, Int32 numRows, Int32 numCols)
void	CopyUpperToLowerColMajor(Double[] squareFullMatrix, Int32 order)
Double[,]	FullColMajorToArray2D(Double[] array1D, Int32 numRows, Int32 numColumns)
Double[]	FullColMajorToFullLowerColMajor(Double[] full, Boolean unitDiagonal)	Extract the lower part only. Both I/O are column major with full storage.
Double[]	FullColMajorToFullLowerRowMajor(Double[] full, Boolean unitDiagonal)	Extract the lower part only. Both I/O are column major with full storage.
Double[]	FullColMajorToFullUpperColMajor(Double[] full, Boolean unitDiagonal)	Extract the upper part only. Both I/O are column major with full storage.
Double[]	FullColMajorToPackedLowerColMajor(Int32 order, Double[] full)	Extract the lower part only. Both I/O are column major with full storage.
void	FullColMajorToPackedLowerColMajor(Int32 order, Double[] full, Double[] packedLower, Int32 offsetLower)	Extract the lower part only. Both I/O are column major with full storage.
Double[]	FullColMajorToPackedUpperColMajor(Int32 order, Double[] full)	Extract the upper part only. Both I/O are column major with full storage.
void	FullColMajorToPackedUpperColMajor(Int32 order, Double[] full, Double[] packedUpper, Int32 offsetUpper)	Extract the upper part only. Both I/O are column major with full storage.
Int32	FullLengthToOrder(Int32 length)
Double[,]	FullLowerColMajorToArray2D(Double[] array1D, Boolean unitDiagonal)
Double[,]	FullRowMajorToArray2D(Double[] array1D, Int32 numRows, Int32 numColumns)
Double[,]	FullUpperColMajorToArray2D(Double[] array1D, Boolean unitDiagonal)
Int32	PackedLengthToOrder(Int32 length)
Double[,]	PackedLowerColMajorToArray2D(Double[] array1D)
Double[,]	PackedLowerRowMajorToArray2D(Double[] array1D)
Double[]	PackedLowerRowMajorToFullColMajor(Double[] packed, Int32 order = 0)
Double[,]	PackedUpperColMajorToArray2D(Double[] array1D, Int32 order = 0)
Double[,]	PackedUpperColMajorToArray2DSymm(Double[] packed, Int32 order = 0)
Double[]	PackedUpperColMajorToFullColMajor(Double[] packed, Int32 order = 0)
Double[]	PackedUpperColMajorToFullSymmColMajor(Double[] packed, Int32 order = 0)	Converts a symmetric matrix from packed column major format to full column major format.
Double[]	PackedUpperRowMajorPackedUpperColMajor(Int32 order, Double[] rowMajor)
Double[,]	PackedUpperRowMajorToArray2D(Double[] array1D)
Double[]	RectColMajorToPackedLowerColMajor(Int32 numRows, Int32 numCols, Double[] rect)
Double[]	RectColMajorToPackedUpperColMajor(Int32 numRows, Int32 numCols, Double[] rect)
Double[]	RectColMajorToRectLowerColMajor(Int32 numRows, Int32 numCols, Double[] rect)	Extract the lower trapezoid (subdiagonal) part only. Both I/O are column major with full storage. Appropriate for rectangular matrices.
Double[]	RectColMajorToRectUpperColMajor(Int32 numRows, Int32 numCols, Double[] rect)	Extract the upper trapezoid (superdiagonal) part only. Both I/O are column major with full storage. Appropriate for rectangular matrices.
Double[]	RectColMajorToSquareFullLowerColMajor(Int32 numRows, Int32 numCols, Double[] rect)
Double[]	RectColMajorToSquareFullUpperColMajor(Int32 numRows, Int32 numCols, Double[] rect)
Double[]	RectColMajorToSquarePackedUpperColMajor(Int32 numRows, Int32 numCols, Double[] rect)

DefiniteProperty

Property of a symmetric matrix. For more see: . Authors: Serafeim Bakalakos

public enum MGroup.LinearAlgebra.Commons.DefiniteProperty
    : Enum, IComparable, IFormattable, IConvertible

Enum

Value	Name	Summary
0	PositiveDefinite	We are sure that the matrix is positive definite.
1	PositiveSemiDefinite	We are sure that the matrix is positive semi-definite (including positive definite).
2	Indefinite	We are sure that the matrix is not positive definite, positive semi-definite, negative definite or negative semi-definite.
3	Unknown	We do not yet know the definiteness of the matrix.

ParallelExtensions

Utility methods for parallel vector operations. Authors: George Stavroulakis

public static class MGroup.LinearAlgebra.Commons.ParallelExtensions

Static Fields

Type	Name	Summary
Int32	AffinityCount

Static Methods

Type	Name	Summary
void	AssignTotalAffinityCount()
void	AssignTotalAffinityCount(Int32 count)
IEnumerable<Tuple<Int32, Int32, Int32>>	PartitionLimits(this T[] vector, Int32 chunks)
IEnumerable<Tuple<Int32, Int32, Int32>>	PartitionLimits(this Double[] vector, Int32 chunks)
IEnumerable<Tuple<Int32, Int32, Int32>>	PartitionLimits(this IVectorView vector, Int32 chunks)

Preconditions

Checks for the compatibility of vector or matrix dimensions for various operations and for the indices. Also templates for exception messages. Authors: Serafeim Bakalakos

public static class MGroup.LinearAlgebra.Commons.Preconditions

Static Methods

Type	Name	Summary
Boolean	AreSameMatrixDimensions(IIndexable2D matrix1, IIndexable2D matrix2)
void	CheckIndex1D(IVectorView vector, Int32 idx)
void	CheckIndexCol(IIndexable2D matrix, Int32 colIdx)
void	CheckIndexRow(IIndexable2D matrix, Int32 rowIdx)
void	CheckIndices(IIndexable2D matrix, Int32 rowIdx, Int32 colIdx)
void	CheckMultiplicationDimensions(IIndexable2D leftMatrix, IIndexable2D rightMatrix)
void	CheckMultiplicationDimensions(Int32 leftColumns, Int32 rightRows)
void	CheckMultiplicationDimensions(IIndexable2D matrix, IVectorView lhsVector, Int32 lhsOffset, IVectorView rhsVector, Int32 rhsOffset, Boolean transposeMatrix)
void	CheckMultiplicationDimensionsSection(IIndexable2D matrixLeft, IVectorView vectorRight, Int32 vectorStart, IVectorView result, Int32 resultStart)
void	CheckSameColDimension(IIndexable2D matrix, IVectorView vector)
void	CheckSameColDimension(IIndexable2D matrix1, IIndexable2D matrix2)
void	CheckSameMatrixDimensions(IIndexable2D matrix1, IIndexable2D matrix2)
void	CheckSameRowDimension(IIndexable2D matrix, IVectorView vector)
void	CheckSameRowDimension(IIndexable2D matrix1, IIndexable2D matrix2)
void	CheckSquare(IIndexable2D matrix)
void	CheckSquare(Int32 numRows, Int32 numColumns)
void	CheckSubvectorDimensions(IIndexable1D vector, Int32 startIndex, Int32 subvectorLength)
void	CheckSystemSolutionDimensions(IIndexable2D matrix, IVectorView rhsVector)
void	CheckSystemSolutionDimensions(Int32 matrixRows, Int32 rhsVectorLength)
void	CheckVectorDimensions(Double[] vector1, Double[] vector2)
void	CheckVectorDimensions(IIndexable1D vector1, IIndexable1D vector2)

ValueComparer

public class MGroup.LinearAlgebra.Commons.ValueComparer

Methods

Type	Name	Summary
Boolean	AreEqual(Double val1, Double val2)