Expression object for sparse matrix-dense vector multiplications.The SMatDVecMultExpr class represents the compile time expression for multiplications between row-major sparse matrices and dense vectors.
More...
#include <SMatDVecMultExpr.h>
Inherits blaze::MatVecMultExpr< DenseVector< SMatDVecMultExpr< MT, VT >, false > >, and blaze::Computation.
|
enum | : bool { simdEnabled = false
} |
| Compilation switch for the expression template evaluation strategy.
|
|
enum | : bool { smpAssignable
} |
| Compilation switch for the expression template assignment strategy.
|
|
using | This = SMatDVecMultExpr< MT, VT > |
| Type of this SMatDVecMultExpr instance.
|
|
using | ResultType = MultTrait_< MRT, VRT > |
| Result type for expression template evaluations.
|
|
using | TransposeType = TransposeType_< ResultType > |
| Transpose type for expression template evaluations.
|
|
using | ElementType = ElementType_< ResultType > |
| Resulting element type.
|
|
using | ReturnType = const ElementType |
| Return type for expression template evaluations.
|
|
using | CompositeType = IfTrue_< useAssign, const ResultType, const SMatDVecMultExpr &> |
| Data type for composite expression templates.
|
|
using | LeftOperand = If_< IsExpression< MT >, const MT, const MT &> |
| Composite type of the left-hand side sparse matrix expression.
|
|
using | RightOperand = If_< IsExpression< VT >, const VT, const VT &> |
| Composite type of the right-hand side dense vector expression.
|
|
using | LT = IfTrue_< evaluateMatrix, const MRT, MCT > |
| Type for the assignment of the left-hand side sparse matrix operand.
|
|
using | RT = IfTrue_< evaluateVector, const VRT, VCT > |
| Type for the assignment of the right-hand side dense matrix operand.
|
|
using | VectorType = VT |
| Type of the vector.
|
|
|
enum | : bool { evaluateMatrix = RequiresEvaluation<MT>::value
} |
| Compilation switch for the composite type of the left-hand side sparse matrix expression.
|
|
enum | : bool { evaluateVector = IsComputation<VT>::value || RequiresEvaluation<VT>::value
} |
| Compilation switch for the composite type of the right-hand side dense vector expression.
|
|
enum | : bool { useAssign = evaluateMatrix || evaluateVector
} |
| Compilation switch for the evaluation strategy of the multiplication expression. More...
|
|
using | MRT = ResultType_< MT > |
| Result type of the left-hand side sparse matrix expression.
|
|
using | VRT = ResultType_< VT > |
| Result type of the right-hand side dense vector expression.
|
|
using | MCT = CompositeType_< MT > |
| Composite type of the left-hand side sparse matrix expression.
|
|
using | VCT = CompositeType_< VT > |
| Composite type of the right-hand side dense vector expression.
|
|
|
LeftOperand | mat_ |
| Left-hand side sparse matrix of the multiplication expression.
|
|
RightOperand | vec_ |
| Right-hand side dense vector of the multiplication expression.
|
|
template<typename MT, typename VT>
class blaze::SMatDVecMultExpr< MT, VT >
Expression object for sparse matrix-dense vector multiplications.
The SMatDVecMultExpr class represents the compile time expression for multiplications between row-major sparse matrices and dense vectors.
◆ anonymous enum
template<typename MT , typename VT >
Compilation switch for the evaluation strategy of the multiplication expression.
The useAssign compile time constant expression represents a compilation switch for the evaluation strategy of the multiplication expression. In case either the matrix or the vector operand requires an intermediate evaluation or the dense vector expression is a compound expression, useAssign will be set to true and the multiplication expression will be evaluated via the assign function family. Otherwise useAssign will be set to false and the expression will be evaluated via the subscript operator.
◆ SMatDVecMultExpr()
template<typename MT , typename VT >
Constructor for the SMatDVecMultExpr class.
- Parameters
-
mat | The left-hand side sparse matrix operand of the multiplication expression. |
vec | The right-hand side dense vector operand of the multiplication expression. |
◆ at()
template<typename MT , typename VT >
Checked access to the vector elements.
- Parameters
-
index | Access index. The index has to be in the range . |
- Returns
- The resulting value.
- Exceptions
-
std::out_of_range | Invalid vector access index. |
◆ canAlias()
template<typename MT , typename VT >
template<typename T >
Returns whether the expression can alias with the given address alias.
- Parameters
-
alias | The alias to be checked. |
- Returns
- true in case the expression can alias, false otherwise.
◆ canSMPAssign()
template<typename MT , typename VT >
Returns whether the expression can be used in SMP assignments.
- Returns
- true in case the expression can be used in SMP assignments, false if not.
◆ isAliased()
template<typename MT , typename VT >
template<typename T >
Returns whether the expression is aliased with the given address alias.
- Parameters
-
alias | The alias to be checked. |
- Returns
- true in case an alias effect is detected, false otherwise.
◆ isAligned()
template<typename MT , typename VT >
Returns whether the operands of the expression are properly aligned in memory.
- Returns
- true in case the operands are aligned, false if not.
◆ leftOperand()
template<typename MT , typename VT >
Returns the left-hand side sparse matrix operand.
- Returns
- The left-hand side sparse matrix operand.
◆ operator[]()
template<typename MT , typename VT >
Subscript operator for the direct access to the vector elements.
- Parameters
-
index | Access index. The index has to be in the range . |
- Returns
- The resulting value.
◆ operator~() [1/2]
template<typename VT, bool TF>
Conversion operator for non-constant vectors.
- Returns
- Reference of the actual type of the vector.
◆ operator~() [2/2]
template<typename VT, bool TF>
Conversion operator for constant vectors.
- Returns
- Const reference of the actual type of the vector.
◆ rightOperand()
template<typename MT , typename VT >
Returns the right-hand side dense vector operand.
- Returns
- The right-hand side dense vector operand.
◆ size()
template<typename MT , typename VT >
Returns the current size/dimension of the vector.
- Returns
- The size of the vector.
The documentation for this class was generated from the following files: