org.apache.commons.math3.random
Class RandomDataGenerator

java.lang.Object
  org.apache.commons.math3.random.RandomDataGenerator

All Implemented Interfaces:

Serializable, RandomData

public class RandomDataGenerator
extends Object
implements RandomData, Serializable

Implements the RandomData interface using a RandomGenerator instance to generate non-secure data and a SecureRandom instance to provide data for the nextSecureXxx methods. If no RandomGenerator is provided in the constructor, the default is to use a Well19937c generator. To plug in a different implementation, either implement RandomGenerator directly or extend AbstractRandomGenerator.

Supports reseeding the underlying pseudo-random number generator (PRNG). The SecurityProvider and Algorithm used by the SecureRandom instance can also be reset.

For details on the default PRNGs, see Random and SecureRandom.

Usage Notes:

• Instance variables are used to maintain RandomGenerator and SecureRandom instances used in data generation. Therefore, to generate a random sequence of values or strings, you should use just one RandomDataImpl instance repeatedly.
• The "secure" methods are *much* slower. These should be used only when a cryptographically secure random sequence is required. A secure random sequence is a sequence of pseudo-random values which, in addition to being well-dispersed (so no subsequence of values is an any more likely than other subsequence of the the same length), also has the additional property that knowledge of values generated up to any point in the sequence does not make it any easier to predict subsequent values.
• When a new RandomDataImpl is created, the underlying random number generators are not initialized. If you do not explicitly seed the default non-secure generator, it is seeded with the current time in milliseconds plus the system identity hash code on first use. The same holds for the secure generator. If you provide a RandomGenerator to the constructor, however, this generator is not reseeded by the constructor nor is it reseeded on first use.
• The reSeed and reSeedSecure methods delegate to the corresponding methods on the underlying RandomGenerator and SecureRandom instances. Therefore, reSeed(long) fully resets the initial state of the non-secure random number generator (so that reseeding with a specific value always results in the same subsequent random sequence); whereas reSeedSecure(long) does not reinitialize the secure random number generator (so secure sequences started with calls to reseedSecure(long) won't be identical).
• This implementation is not synchronized. The underlying RandomGenerator or SecureRandom instances are not protected by synchronization and are not guaranteed to be thread-safe. Therefore, if an instance of this class is concurrently utilized by multiple threads, it is the responsibility of client code to synchronize access to seeding and data generation methods.

Since:

3.1

Version:

$Id: RandomDataGenerator.java 7721 2013-02-14 14:07:13Z CardosoP $

See Also:

Serialized Form

Constructor Summary

RandomDataGenerator()
Construct a RandomDataGenerator, using a default random generator as the source of randomness.

RandomDataGenerator(RandomGenerator rand)
Construct a RandomDataGenerator using the supplied RandomGenerator as the source of (non-secure) random data.

Method Summary

double	nextBeta(double alpha, double beta) Generates a random value from the Beta Distribution.
int	nextBinomial(int numberOfTrials, double probabilityOfSuccess) Generates a random value from the Binomial Distribution.
double	nextCauchy(double median, double scale) Generates a random value from the Cauchy Distribution.
double	nextChiSquare(double df) Generates a random value from the ChiSquare Distribution.
double	nextExponential(double mean) Generates a random value from the exponential distribution with specified mean.
double	nextF(double numeratorDf, double denominatorDf) Generates a random value from the F Distribution.
double	nextGamma(double shape, double scale) Generates a random value from the Gamma Distribution.
double	nextGaussian(double mu, double sigma) Generates a random value from the Normal (or Gaussian) distribution with specified mean and standard deviation.
String	nextHexString(int len) Generates a random string of hex characters of length len.
int	nextHypergeometric(int populationSize, int numberOfSuccesses, int sampleSize) Generates a random value from the Hypergeometric Distribution.
int	nextInt(int lower, int upper) Generates a uniformly distributed random integer between lower and upper (endpoints included).
long	nextLong(long lower, long upper) Generates a uniformly distributed random long integer between lower and upper (endpoints included).
int	nextPascal(int r, double p) Generates a random value from the Pascal Distribution.
int[]	nextPermutation(int n, int k) Generates an integer array of length k whose entries are selected randomly, without repetition, from the integers 0, ..., n - 1 (inclusive).
long	nextPoisson(double mean) Generates a random value from the Poisson distribution with the given mean.
Object[]	nextSample(Collection<?> c, int k) Returns an array of k objects selected randomly from the Collection c.
String	nextSecureHexString(int len) Generates a random string of hex characters from a secure random sequence.
int	nextSecureInt(int lower, int upper) Generates a uniformly distributed random integer between lower and upper (endpoints included) from a secure random sequence.
long	nextSecureLong(long lower, long upper) Generates a uniformly distributed random long integer between lower and upper (endpoints included) from a secure random sequence.
double	nextT(double df) Generates a random value from the T Distribution.
double	nextUniform(double lower, double upper) Generates a uniformly distributed random value from the open interval (lower, upper) (i.e., endpoints excluded).
double	nextUniform(double lower, double upper, boolean lowerInclusive) Generates a uniformly distributed random value from the interval (lower, upper) or the interval [lower, upper).
double	nextWeibull(double shape, double scale) Generates a random value from the Weibull Distribution.
int	nextZipf(int numberOfElements, double exponent) Generates a random value from the Zipf Distribution.
void	reSeed() Reseeds the random number generator with System.currentTimeMillis() + System.identityHashCode(this)).
void	reSeed(long seed) Reseeds the random number generator with the supplied seed.
void	reSeedSecure() Reseeds the secure random number generator with the current time in milliseconds.
void	reSeedSecure(long seed) Reseeds the secure random number generator with the supplied seed.
void	setSecureAlgorithm(String algorithm, String provider) Sets the PRNG algorithm for the underlying SecureRandom instance using the Security Provider API.

Methods inherited from class java.lang.Object

clone, equals, finalize, getClass, hashCode, notify, notifyAll, toString, wait, wait, wait

Constructor Detail

RandomDataGenerator

public RandomDataGenerator()

Construct a RandomDataGenerator, using a default random generator as the source of randomness.

The default generator is a Well19937c seeded with System.currentTimeMillis() + System.identityHashCode(this)). The generator is initialized and seeded on first use.

RandomDataGenerator

public RandomDataGenerator(RandomGenerator rand)

Construct a RandomDataGenerator using the supplied RandomGenerator as the source of (non-secure) random data.

Parameters:

rand - the source of (non-secure) random data (may be null, resulting in the default generator)

Method Detail

nextHexString

public String nextHexString(int len)
                     throws NotStrictlyPositiveException

Generates a random string of hex characters of length len.

The generated string will be random, but not cryptographically secure. To generate cryptographically secure strings, use RandomData.nextSecureHexString(int).

Algorithm Description: hex strings are generated using a 2-step process.

1. len / 2 + 1 binary bytes are generated using the underlying Random
2. Each binary byte is translated into 2 hex digits

Specified by:

nextHexString in interface RandomData

Parameters:

len - the desired string length.

Returns:

the random string.

Throws:

NotStrictlyPositiveException - if len <= 0.

nextInt

public int nextInt(int lower,
                   int upper)
            throws NumberIsTooLargeException

Generates a uniformly distributed random integer between lower and upper (endpoints included).

The generated integer will be random, but not cryptographically secure. To generate cryptographically secure integer sequences, use RandomData.nextSecureInt(int, int).

Specified by:

nextInt in interface RandomData

Parameters:

lower - lower bound for generated integer

upper - upper bound for generated integer

Returns:

a random integer greater than or equal to lower and less than or equal to upper

Throws:

NumberIsTooLargeException - if lower >= upper

nextLong

public long nextLong(long lower,
                     long upper)
              throws NumberIsTooLargeException

Generates a uniformly distributed random long integer between lower and upper (endpoints included).

The generated long integer values will be random, but not cryptographically secure. To generate cryptographically secure sequences of longs, use RandomData.nextSecureLong(long, long).

Specified by:

nextLong in interface RandomData

Parameters:

lower - lower bound for generated long integer

upper - upper bound for generated long integer

Returns:

a random long integer greater than or equal to lower and less than or equal to upper

Throws:

NumberIsTooLargeException - if lower >= upper

nextSecureHexString

public String nextSecureHexString(int len)
                           throws NotStrictlyPositiveException

Generates a random string of hex characters from a secure random sequence.

If cryptographic security is not required, use RandomData.nextHexString(int).

Algorithm Description: hex strings are generated in 40-byte segments using a 3-step process.

1. 20 random bytes are generated using the underlying SecureRandom.
2. SHA-1 hash is applied to yield a 20-byte binary digest.
3. Each byte of the binary digest is converted to 2 hex digits.

Specified by:

nextSecureHexString in interface RandomData

Parameters:

len - the length of the string to be generated

Returns:

a random string of hex characters of length len

Throws:

NotStrictlyPositiveException - if len <= 0

nextSecureInt

public int nextSecureInt(int lower,
                         int upper)
                  throws NumberIsTooLargeException

Generates a uniformly distributed random integer between lower and upper (endpoints included) from a secure random sequence.

Sequences of integers generated using this method will be cryptographically secure. If cryptographic security is not required, RandomData.nextInt(int, int) should be used instead of this method.

Definition: Secure Random Sequence

Specified by:

nextSecureInt in interface RandomData

Parameters:

lower - lower bound for generated integer

upper - upper bound for generated integer

Returns:

a random integer greater than or equal to lower and less than or equal to upper.

Throws:

NumberIsTooLargeException - if lower >= upper.

nextSecureLong

public long nextSecureLong(long lower,
                           long upper)
                    throws NumberIsTooLargeException

Generates a uniformly distributed random long integer between lower and upper (endpoints included) from a secure random sequence.

Sequences of long values generated using this method will be cryptographically secure. If cryptographic security is not required, RandomData.nextLong(long, long) should be used instead of this method.

Definition: Secure Random Sequence

Specified by:

nextSecureLong in interface RandomData

Parameters:

lower - lower bound for generated integer

upper - upper bound for generated integer

Returns:

a random long integer greater than or equal to lower and less than or equal to upper.

Throws:

NumberIsTooLargeException - if lower >= upper.

nextPoisson

public long nextPoisson(double mean)
                 throws NotStrictlyPositiveException

Generates a random value from the Poisson distribution with the given mean.

Definition: Poisson Distribution

Algorithm Description:

• For small means, uses simulation of a Poisson process using Uniform deviates, as described here. The Poisson process (and hence value returned) is bounded by 1000 * mean.
• For large means, uses the rejection algorithm described in
  Devroye, Luc. (1981).The Computer Generation of Poisson Random Variables Computing vol. 26 pp. 197-207.

Specified by:

nextPoisson in interface RandomData

Parameters:

mean - the mean of the Poisson distribution

Returns:

a random value following the specified Poisson distribution

Throws:

NotStrictlyPositiveException - if len <= 0

nextGaussian

public double nextGaussian(double mu,
                           double sigma)
                    throws NotStrictlyPositiveException

Generates a random value from the Normal (or Gaussian) distribution with specified mean and standard deviation.

Definition: Normal Distribution

Specified by:

nextGaussian in interface RandomData

Parameters:

mu - the mean of the distribution

sigma - the standard deviation of the distribution

Returns:

a random value following the specified Gaussian distribution

Throws:

NotStrictlyPositiveException - if sigma <= 0.

nextExponential

public double nextExponential(double mean)
                       throws NotStrictlyPositiveException

Generates a random value from the exponential distribution with specified mean.

Definition: Exponential Distribution

Algorithm Description: Uses the Algorithm SA (Ahrens) from p. 876 in: [1]: Ahrens, J. H. and Dieter, U. (1972). Computer methods for sampling from the exponential and normal distributions. Communications of the ACM, 15, 873-882.

Specified by:

nextExponential in interface RandomData

Parameters:

mean - the mean of the distribution

Returns:

a random value following the specified exponential distribution

Throws:

NotStrictlyPositiveException - if mean <= 0.

nextGamma

public double nextGamma(double shape,
                        double scale)
                 throws NotStrictlyPositiveException

Generates a random value from the Gamma Distribution.

This implementation uses the following algorithms:

For 0 < shape < 1:
Ahrens, J. H. and Dieter, U., Computer methods for sampling from gamma, beta, Poisson and binomial distributions. Computing, 12, 223-246, 1974.

For shape >= 1:
Marsaglia and Tsang, A Simple Method for Generating Gamma Variables. ACM Transactions on Mathematical Software, Volume 26 Issue 3, September, 2000.

Parameters:

shape - the median of the Gamma distribution

scale - the scale parameter of the Gamma distribution

Returns:

random value sampled from the Gamma(shape, scale) distribution

Throws:

NotStrictlyPositiveException - if shape <= 0 or scale <= 0.

nextHypergeometric

public int nextHypergeometric(int populationSize,
                              int numberOfSuccesses,
                              int sampleSize)
                       throws NotPositiveException,
                              NotStrictlyPositiveException,
                              NumberIsTooLargeException

Generates a random value from the Hypergeometric Distribution.

Parameters:

populationSize - the population size of the Hypergeometric distribution

numberOfSuccesses - number of successes in the population of the Hypergeometric distribution

sampleSize - the sample size of the Hypergeometric distribution

Returns:

random value sampled from the Hypergeometric(numberOfSuccesses, sampleSize) distribution

Throws:

NumberIsTooLargeException - if numberOfSuccesses > populationSize, or sampleSize > populationSize.

NotStrictlyPositiveException - if populationSize <= 0.

NotPositiveException - if numberOfSuccesses < 0.

nextPascal

public int nextPascal(int r,
                      double p)
               throws NotStrictlyPositiveException,
                      OutOfRangeException

Generates a random value from the Pascal Distribution.

Parameters:

r - the number of successes of the Pascal distribution

p - the probability of success of the Pascal distribution

Returns:

random value sampled from the Pascal(r, p) distribution

Throws:

NotStrictlyPositiveException - if the number of successes is not positive

OutOfRangeException - if the probability of success is not in the range [0, 1].

nextT

public double nextT(double df)
             throws NotStrictlyPositiveException

Generates a random value from the T Distribution.

Parameters:

df - the degrees of freedom of the T distribution

Returns:

random value from the T(df) distribution

Throws:

NotStrictlyPositiveException - if df <= 0

nextWeibull

public double nextWeibull(double shape,
                          double scale)
                   throws NotStrictlyPositiveException

Generates a random value from the Weibull Distribution.

Parameters:

shape - the shape parameter of the Weibull distribution

scale - the scale parameter of the Weibull distribution

Returns:

random value sampled from the Weibull(shape, size) distribution

Throws:

NotStrictlyPositiveException - if shape <= 0 or scale <= 0.

nextZipf

public int nextZipf(int numberOfElements,
                    double exponent)
             throws NotStrictlyPositiveException

Generates a random value from the Zipf Distribution.

Parameters:

numberOfElements - the number of elements of the ZipfDistribution

exponent - the exponent of the ZipfDistribution

Returns:

random value sampled from the Zipf(numberOfElements, exponent) distribution

Throws:

NotStrictlyPositiveException - if numberOfElements <= 0 or exponent <= 0.

nextBeta

public double nextBeta(double alpha,
                       double beta)

Generates a random value from the Beta Distribution.

Parameters:

alpha - first distribution shape parameter

beta - second distribution shape parameter

Returns:

random value sampled from the beta(alpha, beta) distribution

nextBinomial

public int nextBinomial(int numberOfTrials,
                        double probabilityOfSuccess)

Generates a random value from the Binomial Distribution.

Parameters:

numberOfTrials - number of trials of the Binomial distribution

probabilityOfSuccess - probability of success of the Binomial distribution

Returns:

random value sampled from the Binomial(numberOfTrials, probabilityOfSuccess) distribution

nextCauchy

public double nextCauchy(double median,
                         double scale)

Generates a random value from the Cauchy Distribution.

Parameters:

median - the median of the Cauchy distribution

scale - the scale parameter of the Cauchy distribution

Returns:

random value sampled from the Cauchy(median, scale) distribution

nextChiSquare

public double nextChiSquare(double df)

Generates a random value from the ChiSquare Distribution.

Parameters:

df - the degrees of freedom of the ChiSquare distribution

Returns:

random value sampled from the ChiSquare(df) distribution

nextF

public double nextF(double numeratorDf,
                    double denominatorDf)
             throws NotStrictlyPositiveException

Generates a random value from the F Distribution.

Parameters:

numeratorDf - the numerator degrees of freedom of the F distribution

denominatorDf - the denominator degrees of freedom of the F distribution

Returns:

random value sampled from the F(numeratorDf, denominatorDf) distribution

Throws:

NotStrictlyPositiveException - if numeratorDf <= 0 or denominatorDf <= 0.

nextUniform

public double nextUniform(double lower,
                          double upper)
                   throws NumberIsTooLargeException,
                          NotFiniteNumberException,
                          NotANumberException

Generates a uniformly distributed random value from the open interval (lower, upper) (i.e., endpoints excluded).

Definition: Uniform Distribution lower and upper - lower are the location and scale parameters, respectively.

Algorithm Description: scales the output of Random.nextDouble(), but rejects 0 values (i.e., will generate another random double if Random.nextDouble() returns 0). This is necessary to provide a symmetric output interval (both endpoints excluded).

Specified by:

nextUniform in interface RandomData

Parameters:

lower - the exclusive lower bound of the support

upper - the exclusive upper bound of the support

Returns:

a uniformly distributed random value between lower and upper (exclusive)

Throws:

NumberIsTooLargeException - if lower >= upper

NotFiniteNumberException - if one of the bounds is infinite

NotANumberException - if one of the bounds is NaN

nextUniform

public double nextUniform(double lower,
                          double upper,
                          boolean lowerInclusive)
                   throws NumberIsTooLargeException,
                          NotFiniteNumberException,
                          NotANumberException

Generates a uniformly distributed random value from the interval (lower, upper) or the interval [lower, upper). The lower bound is thus optionally included, while the upper bound is always excluded.

Definition: Uniform Distribution lower and upper - lower are the location and scale parameters, respectively.

Algorithm Description: if the lower bound is excluded, scales the output of Random.nextDouble(), but rejects 0 values (i.e., will generate another random double if Random.nextDouble() returns 0). This is necessary to provide a symmetric output interval (both endpoints excluded).

Specified by:

nextUniform in interface RandomData

Parameters:

lower - the lower bound of the support

upper - the exclusive upper bound of the support

lowerInclusive - true if the lower bound is inclusive

Returns:

uniformly distributed random value in the (lower, upper) interval, if lowerInclusive is false, or in the [lower, upper) interval, if lowerInclusive is true

Throws:

NumberIsTooLargeException - if lower >= upper

NotFiniteNumberException - if one of the bounds is infinite

NotANumberException - if one of the bounds is NaN

nextPermutation

public int[] nextPermutation(int n,
                             int k)
                      throws NumberIsTooLargeException,
                             NotStrictlyPositiveException

Generates an integer array of length k whose entries are selected randomly, without repetition, from the integers 0, ..., n - 1 (inclusive).

Generated arrays represent permutations of n taken k at a time.

Uses a 2-cycle permutation shuffle. The shuffling process is described here.

Specified by:

nextPermutation in interface RandomData

Parameters:

n - the domain of the permutation

k - the size of the permutation

Returns:

a random k-permutation of n, as an array of integers

Throws:

NumberIsTooLargeException - if k > n.

NotStrictlyPositiveException - if k <= 0.

nextSample

public Object[] nextSample(Collection<?> c,
                           int k)
                    throws NumberIsTooLargeException,
                           NotStrictlyPositiveException

Returns an array of k objects selected randomly from the Collection c.

Sampling from c is without replacement; but if c contains identical objects, the sample may include repeats. If all elements of c are distinct, the resulting object array represents a Simple Random Sample of size k from the elements of c.

Algorithm Description: Uses a 2-cycle permutation shuffle to generate a random permutation of c.size() and then returns the elements whose indexes correspond to the elements of the generated permutation. This technique is described, and proven to generate random samples here

Specified by:

nextSample in interface RandomData

Parameters:

c - the collection to be sampled

k - the size of the sample

Returns:

a random sample of k elements from c

Throws:

NumberIsTooLargeException - if k > c.size().

NotStrictlyPositiveException - if k <= 0.

reSeed

public void reSeed(long seed)

Reseeds the random number generator with the supplied seed.

Will create and initialize if null.

Parameters:

seed - the seed value to use

reSeedSecure

public void reSeedSecure()

Reseeds the secure random number generator with the current time in milliseconds.

Will create and initialize if null.

reSeedSecure

public void reSeedSecure(long seed)

Reseeds the secure random number generator with the supplied seed.

Will create and initialize if null.

Parameters:

seed - the seed value to use

reSeed

public void reSeed()

Reseeds the random number generator with System.currentTimeMillis() + System.identityHashCode(this)).

setSecureAlgorithm

public void setSecureAlgorithm(String algorithm,
                               String provider)
                        throws NoSuchAlgorithmException,
                               NoSuchProviderException

Sets the PRNG algorithm for the underlying SecureRandom instance using the Security Provider API. The Security Provider API is defined in Java Cryptography Architecture API Specification & Reference.

USAGE NOTE: This method carries significant overhead and may take several seconds to execute.

Parameters:

algorithm - the name of the PRNG algorithm

provider - the name of the provider

Throws:

NoSuchAlgorithmException - if the specified algorithm is not available

NoSuchProviderException - if the specified provider is not installed