---

CNDLRE Class Reference

The class CNDLRE. More...

#include <DLRE.h>

Inheritance diagram for CNDLRE::

List of all members.

Public Methods
	CNDLRE (double *, long, double, double=0.0, const char *=NIL, const char *=NIL, unsigned long=ULONG_MAX)
	constructor for non-equi-distant x-values.
	CNDLRE (double, double, double, double, double=0.0, const char *=NIL, const char *=NIL, unsigned long=ULONG_MAX)
	constructor for equi-distant x-values.
	CNDLRE (CNParam *)
	FIXME: Undocumented.
	~CNDLRE ()
	destructor.
virtual void	put (double)=0
	This function is called to put a new sample into the statistical evaluation. More...
virtual double	mean () const
	Return the mean value.
virtual double	variance () const
	Return the variance.
virtual unsigned long	trials () const
	Return the number of values used for evaluation.
virtual double	minimum () const
	Return the minimum value evaluated so far.
virtual double	maximum () const
	Return the maximum value evaluated so far.
virtual bool	end () const
	Return whether evaluation has come to an end (TRUE) or not (FALSE).
virtual Phase	status () const
	Return the phase of evaluation.
virtual void	reset ()
	Reset the evaluation.
void	set_base (double b)
	FIXME: Undocumented.
virtual void	change_error (double ne)
	factor for cond. prob.
virtual double	cur_x_lev ()
	change error.
virtual double	p (double xt)
	FIXME: Undocumented.
virtual long	min_index () const
	returns the prob. of xt.
virtual long	max_index () const
	FIXME: Undocumented.
virtual const resultline *	get_result (long)=0
	FIXME: Undocumented.
virtual CNClassDesc	class_desc () const
	Returns the class description (pointer to instance of CNClass) for runtime type information.
virtual bool	is_a (CNClassDesc desc) const
	Type checking.
virtual void	print (ostream &strm=cout) const
	Print/debug output.
virtual void	dump (ostream &strm=cout) const
	Dump object to stream. More...
Static Public Methods
CNDLRE *	cast_from_object (CNObject *obj)
	Safe type cast.
Protected Types
enum	{ NO_INDEX = -3, GREATER = -2, LOWER = -1 }
	returned by get_index.
enum	Reason { OK, MIN, LAST, NO_REASON }
	reason for status = END.
Protected Methods
	CNDLRE ()
	FIXME: Undocumented.
long	get_index (double &)
	FIXME: Undocumented.
Protected Attributes
result *	x
	FIXME: Undocumented.
resultline	line
	array of results.
double	reler_max
	temporary struct holding a single result-line.
unsigned long	nrv
	max. relative error.
unsigned long	max_nrv
	number of trials.
double	min_value
	maximal number of trials.
double	max_value
	minimal value of trials.
long	wasted_left
	maximal value of trials.
long	wasted_right
	values lower minimum.
double	rv_sum
	values greater maximum.
double	rv_square_sum
	necess. for mean.
long	h
	and variance.
long	n0
	run time control (rtc) counter.
long	xoffset
	trials without rtc.
double	xmin
	offset of x-values (step 1).
double	xmax
	minimum of x.
long	index_min
	maximum of x.
long	index_max
	begin of array; normaly 0.
bool	equi_dist
	end of array.
double	int_size
	FIXME: Undocumented.
long	cur_index
	distance between x values.
double	pre_rv
	index of current x-value.
long	pre_index
	predecessor.
double	base
	index of predecessor.
Phase	phase
	needed for conditional probabilties.
bool	end_reached
	state of DLREF.
Reason	reason
	end of evaluation reached ?
long	cur_level_index
	reason for end reached.

---

Detailed Description

The class CNDLRE.

---

Member Function Documentation

virtual void CNDLRE::dump (  ostream &    strm = cout )  const [virtual]

Dump object to stream. Usually more detailed information of the object. Reimplemented from CNObject. Reimplemented in CNDLREP.

virtual void CNDLRE::put (  double    x_i )  [pure virtual]

This function is called to put a new sample into the statistical evaluation. Parameters: val  value of new sample Returns: void Author: Reimplemented from CNStatistics. Reimplemented in CNDLREF, CNDLREG, and CNDLREP.

---

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

• DLRE.h

---