Anasazi::TsqrOrthoManagerImpl< ScalarType, MV > Class Template Reference

TSQR-based OrthoManager subclass implementation. More...

#include <AnasaziTsqrOrthoManager.hpp>

List of all members.

Public Member Functions
	TsqrOrthoManagerImpl (const Teuchos::ParameterList &tsqrParams)
	Constructor.
void	project (MV &X, const_prev_mvs_type Q, prev_coeffs_type C=Teuchos::tuple(serial_matrix_ptr(Teuchos::null))) const
	Compute and .
MagnitudeType	orthonormError (const MV &X) const
	Return .
MagnitudeType	blockReorthogThreshold () const
MagnitudeType	relativeRankTolerance () const

---

Detailed Description

template<class ScalarType, class MV>
class Anasazi::TsqrOrthoManagerImpl< ScalarType, MV >

TSQR-based OrthoManager subclass implementation.

TsqrOrthoManagerImpl implements the interface defined by OrthoManager. It doesn't actually inherit from OrthoManager, which gives us a bit more freedom when defining the actual subclass of OrthoManager (TsqrOrthoManager).

This class uses a combination of Tall Skinny QR (TSQR) and Block Gram-Schmidt (BGS) to orthogonalize multivectors.

The Block Gram-Schmidt procedure used here is inspired by that of G. W. Stewart ("Block Gram-Schmidt Orthogonalization", SISC vol 31 #1 pp. 761--775, 2008), except that we use TSQR+SVD instead of standard Gram-Schmidt with orthogonalization to handle the current block. "Orthogonalization faults" may still happen, but we do not handle them by default. Rather, we make one BGS pass, do TSQR+SVD, check the resulting column norms, and make a second BGS pass (+ TSQR+SVD) if necessary. If we then detect an orthogonalization fault, we throw TsqrOrthoFault.

Definition at line 116 of file AnasaziTsqrOrthoManager.hpp.

---

Constructor & Destructor Documentation

template<class ScalarType , class MV >

Anasazi::TsqrOrthoManagerImpl< ScalarType, MV >::TsqrOrthoManagerImpl

(

const Teuchos::ParameterList &

tsqrParams

)

[inline]

Constructor.

Parameters:

tsqrParams	[in] Configuration parameters for TSQR. See TSQR documentation for how to set those. They depend on which multivector (MV) class you are using (since each MV class maps to a specific TSQR implementation).
Op	[in] Inner product. Don't set to anything not Teuchos::null, otherwise an exception will be thrown. Also, don't call setOp().

Definition at line 140 of file AnasaziTsqrOrthoManager.hpp.

---

Member Function Documentation

template<class ScalarType , class MV >

void Anasazi::TsqrOrthoManagerImpl< ScalarType, MV >::project	(	MV &	X,
		const_prev_mvs_type	Q,
		prev_coeffs_type	C = Teuchos::tuple (serial_matrix_ptr (Teuchos::null))
	)		const [inline]

Compute and .

Warning:

This method does not do reorthogonalization. This is because the reorthogonalization test requires normalization as well.

Definition at line 200 of file AnasaziTsqrOrthoManager.hpp.

template<class ScalarType , class MV >

MagnitudeType Anasazi::TsqrOrthoManagerImpl< ScalarType, MV >::orthonormError

(

const MV &

X

)

const [inline]

Return .

Return the Frobenius norm of I - X^* X, which is an absolute measure of the orthogonality of the columns of X.

Definition at line 497 of file AnasaziTsqrOrthoManager.hpp.

template<class ScalarType , class MV >

MagnitudeType Anasazi::TsqrOrthoManagerImpl< ScalarType, MV >::blockReorthogThreshold

(

)

const [inline]

Relative tolerance for triggering a block reorthogonalization. If any column norm in a block decreases by this amount, then we reorthogonalize.

Definition at line 522 of file AnasaziTsqrOrthoManager.hpp.

template<class ScalarType , class MV >

MagnitudeType Anasazi::TsqrOrthoManagerImpl< ScalarType, MV >::relativeRankTolerance

(

)

const [inline]

Relative tolerance for determining (via the SVD) whether a block is of full numerical rank.

Definition at line 525 of file AnasaziTsqrOrthoManager.hpp.

---

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

• AnasaziTsqrOrthoManager.hpp