2.0 KiB
MatrixNormTriangular
Returns the value of the 1-norm, infinity-norm, Frobenius norm, or the largest absolute value of any element of a trapezoidal m-by-n or triangular matrix. LAPACK function LANTR.
Computing for type matrix
bool matrix::MatrixNormTriangular(
ENUM_BLAS_NORMX norm, // matrix norm
double& norm_value // matrix norm value
);
Computing for type matrix
bool matrixf::MatrixNormTriangular(
ENUM_BLAS_NORMX norm, // matrix norm
float& norm_value // matrix norm value
);
Computing for type matrix
bool matrixc::MatrixNormTriangular(
ENUM_BLAS_NORMX norm, // matrix norm
double& norm_value // matrix norm value
);
Computing for type matrix
bool matrixcf::MatrixNormTriangular(
ENUM_BLAS_NORMX norm, // matrix norm
float& norm_value // matrix norm value
);
Parameters
norm
[in] Value from the ENUM_BLAS_NORMX enumeration, which specifies the value to be returned by the routine.
norm_value
[out] Calculated matrix norm value.
Return Value
Return true if successful, otherwise false in case of an error.
Note
Matrix is triangular if m = n. It can be both upper or lower triangular.
If m < n then matrix must be upper trapezoidal, ie lower m-by-m triangular part must contain zeroes.
If m > n then matrix must be lower trapezoidal, ie upper n-by-n triangular part must contain zeroes.
ENUM_BLAS_NORMX
An enumeration defining which norm calculated.
| ID | Description |
|---|---|
| BLASNORMX_O | 'O': One-norm |
| BLASNORMX_I | 'I': Infinity-norm |
| BLASNORMX_F | 'F': Frobenius-norm |
| BLASNORMX_M | 'M': max(abs(A(i,j))) |