fe::ext::MatrixPower< MATRIX > Class Template Reference

solve B = A^^power, where A is a matrix More...

#include <MatrixPower.h>

Inheritance diagram for fe::ext::MatrixPower< MATRIX >:

Public Member Functions
template<typename T >
void	solve (MATRIX &B, const MATRIX &A, T a_power) const
void	setIterations (U32 iterations)

Private Attributes
MatrixSqrt< MATRIX >	m_matrixSqrt
U32	m_iterations

Detailed Description

template<typename MATRIX>
class fe::ext::MatrixPower< MATRIX >

solve B = A^^power, where A is a matrix

The power can be any arbitrary real number.

Execution time is roughly proportional to the number of set bits in the integer portion of the floating point power and a fixed number of iterations for the fractional part.

The number of iterations used to compute of the fractional portion of the power can be changed. The maximum error after each iteration is half of the previous iteration, starting with one half. The entire integer portion of the power is always computed.

---

The documentation for this class was generated from the following file:

• MatrixPower.h