Keeper Class Reference

This is the class used to store information about the variables that are used in model simulation. More...

#include <keeper.h>

Collaboration diagram for Keeper:

Public Member Functions
	Keeper ()
	This is the default Keeper constructor. More...
	~Keeper ()
	This is the default Keeper destructor. More...
void	keepVariable (double &value, Parameter &attr)
	This function will register a variable in the internal list, along with an initial value. More...
void	deleteParameter (const double &var)
	This function will delete a variable from the internal list. More...
void	changeVariable (const double &pre, double &post)
	This function will replace the value of a variable with a new value. More...
void	setString (const char *str)
	This function will set the text string that is used to describe a variable. More...
void	addString (const char *str)
	This function will add a text string that is used to describe a variable. More...
void	addString (const string str)
	This function will add a text string that is used to describe a variable. More...
void	clearLast ()
	This function will remove the last text string from the list used to describe variables. More...
void	clearAll ()
	This function will remove all of the text strings from the list used to describe variables. More...
void	storeVariables (double likvalue, const DoubleVector &point)
	This function will store the current value of the variables from the optimisation process. More...
void	resetVariables ()
	This function will reset the variables that have been scaled, to be optimised using the Simulated Annealing optimisation algorithm. More...
void	scaleVariables ()
	This function will scale the variables to be optimised, for the Hooke & Jeeves and BFGS optimisation algorithms. More...
void	getOptFlags (IntVector &opt) const
	This function will return a copy of the flags to denote which variables will be optimsised. More...
void	getSwitches (ParameterVector &sw) const
	This function will return a copy of the names of the variables. More...
void	getLowerBounds (DoubleVector &lbs) const
	This function will return a copy of the lower bounds of the variables. More...
void	getUpperBounds (DoubleVector &ubs) const
	This function will return a copy of the upper bounds of the variables. More...
void	getInitialValues (DoubleVector &val) const
	This function will return a copy of the initial value of the variables. More...
void	getScaledValues (DoubleVector &val) const
	This function will return a copy of the scaled value of the variables. More...
void	getCurrentValues (DoubleVector &val) const
	This function will return a copy of the current value of the variables. More...
void	getOptLowerBounds (DoubleVector &lbs) const
	This function will return a copy of the lower bounds of the variables to be optimised. More...
void	getOptUpperBounds (DoubleVector &ubs) const
	This function will return a copy of the upper bounds of the variables to be optimised. More...
void	getOptInitialValues (DoubleVector &val) const
	This function will return a copy of the initial value of the variables to be optimised. More...
void	getOptScaledValues (DoubleVector &val) const
	This function will return a copy of the scaled value of the variables to be optimised. More...
int	numVariables () const
	This function will return the number of variables. More...
int	numOptVariables () const
	This function will return the number of variables to be optimised. More...
void	Update (int pos, double &value)
	This function will update one Keeper variable with a new value. More...
void	Update (const DoubleVector &val)
	This function will update the Keeper variables with new values from a vector. More...
void	Update (const StochasticData *const Stoch)
	This function will update the Keeper variables with new values from StochasticData. More...
void	openPrintFile (const char *const filename)
	This function will open the output file and write header information. More...
void	writeInitialInformation (const LikelihoodPtrVector &likevec)
	This function will write header information about the model parameters to file. More...
void	writeValues (const LikelihoodPtrVector &likevec, int prec)
	This function will write current information about the model parameters to file. More...
void	writeParams (const OptInfoPtrVector &optvec, const char *const filename, int prec, int interrupt)
	This function will write final information about the model parameters to file in a column format (which can then be used as the starting point for a subsequent model run) More...
void	writeBestValues ()
	This function will display information about the best values of the parameters found so far during an optimisation run. More...
void	checkBounds (const LikelihoodPtrVector &likevec) const
	This function will check that the values of the parameters are within the bounds specified in the input file. More...
int	boundsGiven () const
	This function will return the flag used to denote whether the bounds of the parameters have been specified or not. More...
double	getBestLikelihoodScore () const
	This function will return the best likelihood score found so far by the optimisation process. More...

Protected Attributes
AddressKeeperMatrix	address
	This is the AddressKeeperMatrix used to store information about the value and name of the parameters. More...
DoubleVector	initialvalues
	This is the DoubleVector used to store the initial values of the parameters. More...
DoubleVector	scaledvalues
	This is the DoubleVector used to store the scaled values of the parameters. More...
DoubleVector	values
	This is the DoubleVector used to store the values of the parameters. More...
DoubleVector	bestvalues
	This is the DoubleVector used to store the best values of the parameters that the optimisation process has found so far. More...
IntVector	opt
	This is the IntVector used to store information about whether the parameters are to be optimised. More...
StrStack *	stack
	This is the StrStack used to store information about the use of the parameters. More...
ParameterVector	switches
	This is the ParameterVector used to store information about the names of the parameters. More...
DoubleVector	lowerbds
	This is the DoubleVector used to store information about the lower bounds of the parameters. More...
DoubleVector	upperbds
	This is the DoubleVector used to store information about the upper bounds of the parameters. More...
int	numoptvar
	This is the number of parameters to be optimised. More...
int	boundsgiven
	This is the flag used to denote whether the bounds of the parameters have been specified or not. More...
double	bestlikelihood
	This is the best likelihood score found so far by the optimisation process. More...
int	fileopen
	This is the flag used to denote whether an output file has been specified or not. More...
ofstream	outfile
	This ofstream is the file that all the parameter information gets sent to. More...

Detailed Description

This is the class used to store information about the variables that are used in model simulation.

This class stores all the information about the variables that are used in the model simulation. This includes the current value, the initial value and best value found so far by any optimisation routines, the upper and lower bounds for the variables and a flag to denote whether the variable is to be optimised or not. This class also updates the model parameters with new values for the variables, either as a result of the optimisation routines or when new values for the parameters are received from the network if the simulation is part of a parallel optimisation, using paramin.

Constructor & Destructor Documentation

Keeper()

Keeper::Keeper

(

)

This is the default Keeper constructor.

~Keeper()

Keeper::~Keeper

(

)

This is the default Keeper destructor.

Member Function Documentation

addString() [1/2]

void Keeper::addString

(

const char *

str

)

This function will add a text string that is used to describe a variable.

Parameters

str	is the text string to be stored

addString() [2/2]

void Keeper::addString

(

const string

str

)

This function will add a text string that is used to describe a variable.

Parameters

str	is the text string to be stored

boundsGiven()

int Keeper::boundsGiven ( ) const

inline

This function will return the flag used to denote whether the bounds of the parameters have been specified or not.

Returns

flag to denote whether the bounds have been given or not

changeVariable()

void Keeper::changeVariable	(	const double &	pre,
		double &	post
	)

This function will replace the value of a variable with a new value.

Parameters

pre	is the old value for the variable
post	is the new value for the variable

checkBounds()

void Keeper::checkBounds

(

const LikelihoodPtrVector &

likevec

)

const

This function will check that the values of the parameters are within the bounds specified in the input file.

Parameters

likevec

is the LikelihoodPtrVector containing the likelihood components for the current model

clearAll()

void Keeper::clearAll

(

)

This function will remove all of the text strings from the list used to describe variables.

clearLast()

void Keeper::clearLast

(

)

This function will remove the last text string from the list used to describe variables.

deleteParameter()

void Keeper::deleteParameter

(

const double &

var

)

This function will delete a variable from the internal list.

Parameters

var	is the value for the variable

getBestLikelihoodScore()

double Keeper::getBestLikelihoodScore ( ) const

inline

This function will return the best likelihood score found so far by the optimisation process.

Returns

bestlikelihood

getCurrentValues()

void Keeper::getCurrentValues

(

DoubleVector &

val

)

const

This function will return a copy of the current value of the variables.

Parameters

val	is the DoubleVector that will contain a copy of the current values

getInitialValues()

void Keeper::getInitialValues

(

DoubleVector &

val

)

const

This function will return a copy of the initial value of the variables.

Parameters

val	is the DoubleVector that will contain a copy of the initial values

getLowerBounds()

void Keeper::getLowerBounds

(

DoubleVector &

lbs

)

const

This function will return a copy of the lower bounds of the variables.

Parameters

lbs	is the DoubleVector that will contain a copy of the lower bounds

getOptFlags()

void Keeper::getOptFlags

(

IntVector &

opt

)

const

This function will return a copy of the flags to denote which variables will be optimsised.

Parameters

opt	is the IntVector that will contain a copy of the flags

getOptInitialValues()

void Keeper::getOptInitialValues

(

DoubleVector &

val

)

const

This function will return a copy of the initial value of the variables to be optimised.

Parameters

val	is the DoubleVector that will contain a copy of the initial values

getOptLowerBounds()

void Keeper::getOptLowerBounds

(

DoubleVector &

lbs

)

const

This function will return a copy of the lower bounds of the variables to be optimised.

Parameters

lbs	is the DoubleVector that will contain a copy of the lower bounds

getOptScaledValues()

void Keeper::getOptScaledValues

(

DoubleVector &

val

)

const

This function will return a copy of the scaled value of the variables to be optimised.

Parameters

val	is the DoubleVector that will contain a copy of the scaled values

getOptUpperBounds()

void Keeper::getOptUpperBounds

(

DoubleVector &

ubs

)

const

This function will return a copy of the upper bounds of the variables to be optimised.

Parameters

ubs	is the DoubleVector that will contain a copy of the upper bounds

getScaledValues()

void Keeper::getScaledValues

(

DoubleVector &

val

)

const

This function will return a copy of the scaled value of the variables.

Parameters

val	is the DoubleVector that will contain a copy of the scaled values

getSwitches()

void Keeper::getSwitches

(

ParameterVector &

sw

)

const

This function will return a copy of the names of the variables.

Parameters

sw	is the ParameterVector that will contain a copy of the names

getUpperBounds()

void Keeper::getUpperBounds

(

DoubleVector &

ubs

)

const

This function will return a copy of the upper bounds of the variables.

Parameters

ubs	is the DoubleVector that will contain a copy of the upper bounds

keepVariable()

void Keeper::keepVariable	(	double &	value,
		Parameter &	attr
	)

This function will register a variable in the internal list, along with an initial value.

Parameters

value	is the value for the variable
attr	is the name of the Parameter for the variable

numOptVariables()

int Keeper::numOptVariables ( ) const

inline

This function will return the number of variables to be optimised.

Returns

number of variables to be optimised

numVariables()

int Keeper::numVariables

(

)

const

This function will return the number of variables.

Returns

number of variables

openPrintFile()

void Keeper::openPrintFile

(

const char *const

filename

)

This function will open the output file and write header information.

Parameters

filename

is the name of the file to write the model information to

resetVariables()

void Keeper::resetVariables

(

)

This function will reset the variables that have been scaled, to be optimised using the Simulated Annealing optimisation algorithm.

scaleVariables()

void Keeper::scaleVariables

(

)

This function will scale the variables to be optimised, for the Hooke & Jeeves and BFGS optimisation algorithms.

setString()

void Keeper::setString

(

const char *

str

)

This function will set the text string that is used to describe a variable.

Parameters

str	is the text string to be stored

storeVariables()

void Keeper::storeVariables	(	double	likvalue,
		const DoubleVector &	point
	)

This function will store the current value of the variables from the optimisation process.

Parameters

likvalue	is the current likelihood value
point	is the DoubleVector of the current parameter values

Update() [1/3]

void Keeper::Update

(

const DoubleVector &

val

)

This function will update the Keeper variables with new values from a vector.

Parameters

val	is the DoubleVector containing the new values of the variables

Update() [2/3]

void Keeper::Update

(

const StochasticData *const

Stoch

)

This function will update the Keeper variables with new values from StochasticData.

Parameters

Stoch

is the StochasticData containing the new values of the variables

Update() [3/3]

void Keeper::Update	(	int	pos,
		double &	value
	)

This function will update one Keeper variable with a new value.

Parameters

pos	is the identifier of the variable to update
value	is the new value of the variable

writeBestValues()

void Keeper::writeBestValues

(

)

This function will display information about the best values of the parameters found so far during an optimisation run.

writeInitialInformation()

void Keeper::writeInitialInformation

(

const LikelihoodPtrVector &

likevec

)

This function will write header information about the model parameters to file.

Parameters

likevec

is the LikelihoodPtrVector containing the likelihood components for the current model

writeParams()

void Keeper::writeParams	(	const OptInfoPtrVector &	optvec,
		const char *const	filename,
		int	prec,
		int	interrupt
	)

This function will write final information about the model parameters to file in a column format (which can then be used as the starting point for a subsequent model run)

Parameters

optvec	is the OptInfoPtrVector containing the optimisation algorithms used for the current model
filename	is the name of the file to write the model information to
prec	is the precision to use in the output file
interrupt	is the flag to denote whether the current run was interrupted by the user or not

writeValues()

void Keeper::writeValues	(	const LikelihoodPtrVector &	likevec,
		int	prec
	)

This function will write current information about the model parameters to file.

Parameters

likevec	is the LikelihoodPtrVector containing the likelihood components for the current model
prec	is the precision to use in the output file

Member Data Documentation

address

AddressKeeperMatrix Keeper::address

protected

This is the AddressKeeperMatrix used to store information about the value and name of the parameters.

bestlikelihood

double Keeper::bestlikelihood

protected

This is the best likelihood score found so far by the optimisation process.

bestvalues

DoubleVector Keeper::bestvalues

protected

This is the DoubleVector used to store the best values of the parameters that the optimisation process has found so far.

boundsgiven

int Keeper::boundsgiven

protected

This is the flag used to denote whether the bounds of the parameters have been specified or not.

fileopen

int Keeper::fileopen

protected

This is the flag used to denote whether an output file has been specified or not.

initialvalues

DoubleVector Keeper::initialvalues

protected

This is the DoubleVector used to store the initial values of the parameters.

lowerbds

DoubleVector Keeper::lowerbds

protected

This is the DoubleVector used to store information about the lower bounds of the parameters.

numoptvar

int Keeper::numoptvar

protected

This is the number of parameters to be optimised.

opt

IntVector Keeper::opt

protected

This is the IntVector used to store information about whether the parameters are to be optimised.

Note

If opt[i] is 1 then parameter i is to be optimised, else if opt[i] is 0 then parameter i is not to be optimised and its value is fixed throughout the optimisation process

outfile

ofstream Keeper::outfile

protected

This ofstream is the file that all the parameter information gets sent to.

scaledvalues

DoubleVector Keeper::scaledvalues

protected

This is the DoubleVector used to store the scaled values of the parameters.

stack

StrStack* Keeper::stack

protected

This is the StrStack used to store information about the use of the parameters.

switches

ParameterVector Keeper::switches

protected

This is the ParameterVector used to store information about the names of the parameters.

upperbds

DoubleVector Keeper::upperbds

protected

This is the DoubleVector used to store information about the upper bounds of the parameters.

values

DoubleVector Keeper::values

protected

This is the DoubleVector used to store the values of the parameters.

---

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

• src/include/keeper.h
• src/keeper.cc