Parameterizes a Function by one or more Scalar ordinates. More...

#include <ChainRule.h>

Public Types
typedef T::Result	Result

Public Member Functions
	ChainRule ()
	Default constructor.
	ChainRule (const ChainRule &rule)
	Copy constructor.
ChainRule &	operator= (const ChainRule &rule)
	Assignment operator.
void	set_model (T *model)
	Set the Function to be constrained by Scalar ordinates.
T *	get_model ()
void	set_constraint (unsigned iparam, Scalar *scalar)
	Set the Scalar instance used to constrain the specified parameter.
bool	has_constraint (unsigned iparam)
Scalar *	get_constraint (unsigned iparam)
std::string	get_name () const
	Return the name of the class.

Protected Member Functions
void	calculate (Result &result, std::vector< Result > *gradient)
	Return the Result and its gradient.
ConstrainedParameter *	find_constraint (unsigned iparam)
	Find the matching constraint.

Protected Attributes
std::vector< ConstrainedParameter >	constraints
	Scalars used to constrain the parameters.
Project< T >	model
	The Function to be constrained by Scalar ordinates.

Detailed Description

template<class T>
class MEAL::ChainRule< T >

Parameterizes a Function by one or more Scalar ordinates.

Given any Function, $M({\bf a})$ , one or more parameters may be constrained by a Scalar, $f({\bf b})$ . That is, $a_i=f({\bf b})$ . The fit flag for $a_i$ is set to false, and the partial derivatives of $ M $ with respect to the Scalar parameters, $\bf b$ , are given by the chain rule, ${\partial M\over\partial b_k} = {\partial M\over\partial a_i}{\partial f\over\partial b_k}$ .

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

ChainRule.h

Public Types

Public Member Functions

Protected Member Functions

Protected Attributes

Detailed Description