EGS_ScoringSingle Class Reference

A class for scoring a single quantity of interest in a Monte Carlo simulation. More...

#include <egs_scoring.h>

Public Member Functions
	EGS_ScoringSingle ()
	Construct a scoring object initialized to zero.
void	score (unsigned short ncase, EGS_Float f)
	Add f to the score of the ncase'th statistically independent event. More...
void	finishCase (unsigned short new_case, EGS_Float new_result)
	Finish the current 'case' (event) and start a new event with index new_case and a score of new_result.
EGS_Float	currentScore () const
	Returns the score of the current event.
void	currentScore (EGS_Float &s, unsigned short &ncase) const
	Sets s to the score of the current event and ncase to the index of the current event.
void	currentScore (double &s, double &s2)
	Sets s to the sum of scores collected so far and s2 to the sum of scores squared.
void	currentResult (EGS_I64 ncase, double &r, double &dr)
	Sets r to the current result and dr to its statistical uncertainty assuming ncase statistically independent events. More...
bool	storeState (ostream &data)
	Stores the state of the scoring object into the data stream data. Returns true on success, false on failure. More...
bool	setState (istream &data)
	Set the state of the scoring object from the data stream data. More...
void	reset ()
	Reset the scoring object to a pristine state (i.e. all counters set to zero).
EGS_ScoringSingle &	operator+= (const EGS_ScoringSingle &x)
	Combine the results of two scoring objects. More...

Protected Attributes
double	sum
double	sum2
EGS_Float	tmp
unsigned short	current_ncase

Detailed Description

A class for scoring a single quantity of interest in a Monte Carlo simulation.

This class implements most of the functionality needed for scoring a single quantity of interest in a Monte Carlo simulation, including a history-by-history statistical analysis. However, as this class is meant to be used by EGS_ScoringArray for collecting results of potentially a large number of quantities (e.g. a 3D dose distribution in a XYZ patient geometry), the history counter is represented with an unsigned 16 bit integer and therefore the maximum number of different histories is

1. Hence, for real applications, it is easier to use the EGS_ScoringArray class even for a single quantity of interest.

Definition at line 63 of file egs_scoring.h.

Member Function Documentation

void EGS_ScoringSingle::score	(	unsigned short	ncase,
		EGS_Float	f
	)

Add f to the score of the ncase'th statistically independent event.

If ncase is the same as the ncase passed in the last call to this function, then f is simply added to the score of the current event, otherwise a new statistically independent event is started.

Definition at line 78 of file egs_scoring.h.

void EGS_ScoringSingle::currentResult	(	EGS_I64	ncase,
		double &	r,
		double &	dr
	)

Sets r to the current result and dr to its statistical uncertainty assuming ncase statistically independent events.

This function sets r to the ratio of the sum of scores collected so far and ncase and dr to the statistical uncertainty of r.

Definition at line 122 of file egs_scoring.h.

Referenced by EGS_ScoringArray::reportResults().

bool EGS_ScoringSingle::storeState

(

ostream &

data

)

Stores the state of the scoring object into the data stream data. Returns true on success, false on failure.

This function can be used for storing intermediate results into a data file for later recovery in e.g. restarted calculations. The data stored is the sum of scores, sum of scores squared and the 16 bit integer representing the last statistically independent event that contributed to the score.

See Also

setState().

Definition at line 144 of file egs_scoring.h.

bool EGS_ScoringSingle::setState

(

istream &

data

)

Set the state of the scoring object from the data stream data.

The data extracted from the stream is the 16 bit integer representing the last statistically independent event that contributed to the score, the sum of scores and the sum of scores squared (both double precision).

See Also

storeState()

Definition at line 161 of file egs_scoring.h.

EGS_ScoringSingle& EGS_ScoringSingle::operator+=

(

const EGS_ScoringSingle &

x

)

Combine the results of two scoring objects.

This operator adds the results of the scoring object x to the results collected so far by the invoking object, assuming that x is statistically independent. This method is useful for combining parallel runs.

Definition at line 184 of file egs_scoring.h.

References sum, sum2, and tmp.

Member Data Documentation

double EGS_ScoringSingle::sum

protected

Sum of scores collected so far by the scoring object.

Definition at line 190 of file egs_scoring.h.

Referenced by operator+=().

double EGS_ScoringSingle::sum2

protected

Sum of scores squared collected so far by the scoring object.

Definition at line 198 of file egs_scoring.h.

Referenced by operator+=().

EGS_Float EGS_ScoringSingle::tmp

protected

The score of the current event.

Definition at line 200 of file egs_scoring.h.

Referenced by operator+=().

unsigned short EGS_ScoringSingle::current_ncase

protected

The index of the current statistically independent event

Definition at line 202 of file egs_scoring.h.

---

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

• /home/rwtownson/EGSnrc/HEN_HOUSE/egs++/egs_scoring.h