C# Class PeterO.RandomGenerator

A class that adapts a random byte generator to generate random numbers in a variety of statistical distributions.

The method descriptions in this class assume the underlying random byte generator generates uniformly distributed numbers that are independent of each other.

Thread safety: The methods in this class are safe for concurrent use by multiple threads, as long as the underlying random byte generator is as well.

Show file Open project: peteroupc/CBOR Class Usage Examples

Public Methods

Method	Description
Bernoulli ( ) : bool	Returns either true or false at a 50% chance each.
Bernoulli ( double p ) : bool	Returns either true or false, depending on the given probability.
Binomial ( int trials ) : int	Conceptually, generates either 1 or 0 the given number of times, where either number is equally likely, and counts the number of 1's generated.
Binomial ( int trials, double p ) : int	Conceptually, generates either 1 or 0 the given number of times, where a 1 is generated at the given probability, and counts the number of 1's generated.
ChiSquared ( int df ) : double	Generates a random number that is the sum of the squares of "df" normally-distributed random numbers with a mean of 0 and a standard deviation of 1.
Exponential ( ) : double	Not documented yet.
Gamma ( double a ) : double	Not documented yet.
Gamma ( double a, double b ) : double	Not documented yet.
Geometric ( ) : int	Conceptually, generates either 1 or 0 until a 1 is generated, and counts the number of 0's generated. Either number has an equal probability of being generated.
Geometric ( double p ) : int	Conceptually, generates either 1 or 0 until a 1 is generated, and counts the number of 0's generated. A 1 is generated at the given probability.
Hypergeometric ( int trials, int ones, int count ) : int	Conceptually, given a set of tokens, some of which are labeled 1 and the others labeled 0, draws "trials" tokens at random without replacement and then counts the number of 1's drawn.
LogNormal ( double mean, double sd ) : double	Generates a logarithmic normally-distributed number with the given mean and standard deviation.
NegativeBinomial ( int trials ) : int	Conceptually, generates either 1 or 0 the given number of times until the given number of 1's are generated, and counts the number of 0's generated. Either number has an equal probability of being generated.
NegativeBinomial ( int trials, double p ) : int	Conceptually, generates either 1 or 0 until the given number of 1's are generated, and counts the number of 0's generated. A 1 is generated at the given probability.
Normal ( ) : double	Generates a normally-distributed number with mean 0 and standard deviation 1.
Normal ( double mean, double sd ) : double	Generates a normally-distributed number with the given mean and standard deviation.
Poisson ( double mean ) : int	Generates a random integer such that the average of random numbers approaches the given mean number when this method is called repeatedly with the same mean.
RandomGenerator ( ) : System	Initializes a new instance of the RandomGenerator class.
RandomGenerator ( IRandomGen valueIrg ) : System	Initializes a new instance of the RandomGenerator class.
Uniform ( ) : double	Returns a uniformly-distributed 64-bit floating-point number from 0 and up, but less than 1.
Uniform ( double max ) : double	Returns a uniformly-distributed 64-bit floating-point number from 0 and up, but less than the given number.
Uniform ( double min, double max ) : double	Not documented yet.
UniformInt ( int maxExclusive ) : int	Generates a random 32-bit signed integer 0 or greater and less than the given number.
UniformInt ( int minInclusive, int maxExclusive ) : int	Generates a random 32-bit signed integer within a given range.
UniformLong ( long maxExclusive ) : long	Generates a random 32-bit signed integer 0 or greater and less than the given number.
UniformLong ( long minInclusive, long maxExclusive ) : long	Generates a random 64-bit signed integer within a given range.
UniformSingle ( ) : double	Returns a uniformly-distributed 32-bit floating-point number from 0 and up, but less than 1.

Private Methods

Method	Description
NonZeroUniform ( ) : double

Method Details

Bernoulli() public method

Returns either true or false at a 50% chance each.

public Bernoulli ( ) : bool
return	bool

Bernoulli() public method

Returns either true or false, depending on the given probability.

public Bernoulli ( double p ) : bool
p	double	A probability from 0 through 1. 0 means always /// false, and 1 means always true.
return	bool

Binomial() public method

Conceptually, generates either 1 or 0 the given number of times, where either number is equally likely, and counts the number of 1's generated.

public Binomial ( int trials ) : int
trials	int	The number of times to generate a random /// number, conceptually.
return	int

Binomial() public method

Conceptually, generates either 1 or 0 the given number of times, where a 1 is generated at the given probability, and counts the number of 1's generated.

public Binomial ( int trials, double p ) : int
trials	int	The number of times to generate a random /// number, conceptually.
p	double	The probability for each trial to succeed, from 0 /// (never) to 1 (always).
return	int

ChiSquared() public method

Generates a random number that is the sum of the squares of "df" normally-distributed random numbers with a mean of 0 and a standard deviation of 1.

public ChiSquared ( int df ) : double
df	int	Degrees of freedom (the number of independently /// chosen normally-distributed numbers).
return	double

Exponential() public method

Not documented yet.

public Exponential ( ) : double
return	double

Gamma() public method

Not documented yet.

public Gamma ( double a ) : double
a	double	Another 64-bit floating-point number.
return	double

Gamma() public method

Not documented yet.

public Gamma ( double a, double b ) : double
a	double	Another 64-bit floating-point number.
b	double	A 64-bit floating-point number. (3).
return	double

Geometric() public method

Conceptually, generates either 1 or 0 until a 1 is generated, and counts the number of 0's generated. Either number has an equal probability of being generated.

public Geometric ( ) : int
return	int

Geometric() public method

Conceptually, generates either 1 or 0 until a 1 is generated, and counts the number of 0's generated. A 1 is generated at the given probability.

public Geometric ( double p ) : int
p	double	A 64-bit floating-point number.
return	int

Hypergeometric() public method

Conceptually, given a set of tokens, some of which are labeled 1 and the others labeled 0, draws "trials" tokens at random without replacement and then counts the number of 1's drawn.

public Hypergeometric ( int trials, int ones, int count ) : int
trials	int	The number of tokens drawn at random without /// replacement.
ones	int	The number of tokens labeled 1.
count	int	The number of tokens labeled 1 or 0.
return	int

LogNormal() public method

Generates a logarithmic normally-distributed number with the given mean and standard deviation.

public LogNormal ( double mean, double sd ) : double
mean	double	The desired mean.
sd	double	Standard deviation.
return	double

NegativeBinomial() public method

Conceptually, generates either 1 or 0 the given number of times until the given number of 1's are generated, and counts the number of 0's generated. Either number has an equal probability of being generated.

public NegativeBinomial ( int trials ) : int
trials	int	The number of 1's to generate before the /// process stops.
return	int

NegativeBinomial() public method

Conceptually, generates either 1 or 0 until the given number of 1's are generated, and counts the number of 0's generated. A 1 is generated at the given probability.

public NegativeBinomial ( int trials, double p ) : int
trials	int	The number of 1's to generate before the /// process stops.
p	double	The probability for each trial to succeed, from 0 /// (never) to 1 (always).
return	int

Normal() public method

Generates a normally-distributed number with mean 0 and standard deviation 1.

public Normal ( ) : double
return	double

Normal() public method

Generates a normally-distributed number with the given mean and standard deviation.

public Normal ( double mean, double sd ) : double
mean	double	The desired mean.
sd	double	Standard deviation.
return	double

Poisson() public method

Generates a random integer such that the average of random numbers approaches the given mean number when this method is called repeatedly with the same mean.

public Poisson ( double mean ) : int
mean	double	The expected mean of the random numbers.
return	int

RandomGenerator() public method

Initializes a new instance of the RandomGenerator class.

public RandomGenerator ( ) : System
return	System

RandomGenerator() public method

Initializes a new instance of the RandomGenerator class.

public RandomGenerator ( IRandomGen valueIrg ) : System
valueIrg	IRandomGen	An IRandomGen object.
return	System

Uniform() public method

Returns a uniformly-distributed 64-bit floating-point number from 0 and up, but less than 1.

public Uniform ( ) : double
return	double

Uniform() public method

Returns a uniformly-distributed 64-bit floating-point number from 0 and up, but less than the given number.

public Uniform ( double max ) : double
max	double	Number that the randomly-generated number will be /// less than.
return	double

Uniform() public method

Not documented yet.

public Uniform ( double min, double max ) : double
min	double	Smallest possible number that will be /// generated.
max	double	Number that the randomly-generated number will be /// less than.
return	double

UniformInt() public method

Generates a random 32-bit signed integer 0 or greater and less than the given number.

public UniformInt ( int maxExclusive ) : int
maxExclusive	int	One plus the largest possible value of /// the random number.
return	int

UniformInt() public method

Generates a random 32-bit signed integer within a given range.

public UniformInt ( int minInclusive, int maxExclusive ) : int
minInclusive	int	Smallest possible value of the random /// number.
maxExclusive	int	One plus the largest possible value of /// the random number.
return	int

UniformLong() public method

Generates a random 32-bit signed integer 0 or greater and less than the given number.

public UniformLong ( long maxExclusive ) : long
maxExclusive	long	One plus the largest possible value of /// the random number.
return	long

UniformLong() public method

Generates a random 64-bit signed integer within a given range.

public UniformLong ( long minInclusive, long maxExclusive ) : long
minInclusive	long	Smallest possible value of the random /// number.
maxExclusive	long	One plus the largest possible value of /// the random number.
return	long

UniformSingle() public method

Returns a uniformly-distributed 32-bit floating-point number from 0 and up, but less than 1.

public UniformSingle ( ) : double
return	double