PyTrilinos.NOX.Epetra.InvOperator Class Reference

Inheritance diagram for PyTrilinos.NOX.Epetra.InvOperator:

Collaboration diagram for PyTrilinos.NOX.Epetra.InvOperator:

List of all members.

Public Member Functions
def	__init__
def	SetUseTranspose
def	Apply
def	ApplyInverse
def	NormInf
def	Label
def	Operator
def	UseTranspose
def	HasNormInf
def	Comm
def	OperatorDomainMap
def	OperatorRangeMap
def	__disown__
Public Attributes
	this

---

Detailed Description

Proxy of C++ Epetra_InvOperator class

---

Constructor & Destructor Documentation

def PyTrilinos.NOX.Epetra.InvOperator.__init__	(		self,
			args
	)

__init__(Epetra_InvOperator self, Operator operatorIn) -> InvOperator

Reimplemented from PyTrilinos.NOX.Epetra.Operator.

---

Member Function Documentation

def PyTrilinos.NOX.Epetra.InvOperator.Apply	(		self,
			args
	)

Apply(self, MultiVector x, MultiVector y) -> int

In C++, the Apply() method is pure virtual, thus intended to be
overridden by derived classes.  In python, cross-language polymorphism
is supported, and you are expected to derive classes from this base
class and redefine the Apply() method.  C++ code (e.g., AztecOO
solvers) can call back to your Apply() method as needed.  You must
support two arguments, labeled here MultiVector x and MultiVector y.
These will be converted from Epetra_MultiVector C++ objects to
numpy-hybrid Epetra.MultiVector objects before they are passed to you.
Thus, it is legal to use slice indexing and other numpy features to
compute y from x.

If application of your operator is successful, return 0; else return
some non-zero error code.

It is strongly suggested that you prevent Apply() from raising any
exceptions.  Accidental errors can be prevented by wrapping your code
in a try block:

    try:
# Your code goes here...
    except Exception, e:
print 'A python exception was raised by method Apply:'
print e
return -1

By returning a -1, you inform the calling routine that Apply() was
unsuccessful.

Reimplemented from PyTrilinos.NOX.Epetra.Operator.

def PyTrilinos.NOX.Epetra.InvOperator.ApplyInverse	(		self,
			args
	)

ApplyInverse(self, MultiVector x, MultiVector y) -> int

In C++, the ApplyInverse() method is pure virtual, thus intended to be
overridden by derived classes.  In python, cross-language polymorphism
is supported, and you are expected to derive classes from this base
class and redefine the ApplyInverse() method.  C++ code (e.g., AztecOO
solvers) can call back to your ApplyInverse() method as needed.  You
must support two arguments, labeled here MultiVector x and MultiVector
y.  These will be converted from Epetra_MultiVector C++ objects to
numpy-hybrid Epetra.MultiVector objects before they are passed to you.
Thus, it is legal to use slice indexing and other numpy features to
compute y from x.

If application of your operator is successful, return 0; else return
some non-zero error code.

It is strongly suggested that you prevent ApplyInverse() from raising
any exceptions.  Accidental errors can be prevented by wrapping your
code in a try block:

    try:
# Your code goes here...
    except Exception, e:
print 'A python exception was raised by method ApplyInverse:'
print e
return -1

By returning a -1, you inform the calling routine that ApplyInverse()
was unsuccessful.

Reimplemented from PyTrilinos.NOX.Epetra.Operator.

def PyTrilinos.NOX.Epetra.InvOperator.Comm	(		self,
			args
	)

Comm(InvOperator self) -> Comm

Reimplemented from PyTrilinos.NOX.Epetra.Operator.

def PyTrilinos.NOX.Epetra.InvOperator.HasNormInf	(		self,
			args
	)

HasNormInf(InvOperator self) -> bool

Reimplemented from PyTrilinos.NOX.Epetra.Operator.

def PyTrilinos.NOX.Epetra.InvOperator.Label	(		self,
			args
	)

Label(InvOperator self) -> char const *

Reimplemented from PyTrilinos.NOX.Epetra.Operator.

def PyTrilinos.NOX.Epetra.InvOperator.NormInf	(		self,
			args
	)

NormInf(InvOperator self) -> double

Reimplemented from PyTrilinos.NOX.Epetra.Operator.

def PyTrilinos.NOX.Epetra.InvOperator.Operator	(		self,
			args
	)

Operator(InvOperator self) -> Operator

def PyTrilinos.NOX.Epetra.InvOperator.OperatorDomainMap	(		self,
			args
	)

OperatorDomainMap(InvOperator self) -> Map

Reimplemented from PyTrilinos.NOX.Epetra.Operator.

def PyTrilinos.NOX.Epetra.InvOperator.OperatorRangeMap	(		self,
			args
	)

OperatorRangeMap(InvOperator self) -> Map

Reimplemented from PyTrilinos.NOX.Epetra.Operator.

def PyTrilinos.NOX.Epetra.InvOperator.SetUseTranspose	(		self,
			args
	)

SetUseTranspose(InvOperator self, bool UseTranspose_in) -> int

Reimplemented from PyTrilinos.NOX.Epetra.Operator.

def PyTrilinos.NOX.Epetra.InvOperator.UseTranspose	(		self,
			args
	)

UseTranspose(InvOperator self) -> bool

Reimplemented from PyTrilinos.NOX.Epetra.Operator.

---

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

• NOX/Epetra/__init__.py