MeteoIODoc 20241221.207bde49
mio::RandomNumberGenerator Class Reference

Detailed Description

Author
Michael Reisecker
Date
2018-10

Random Number Generator

Random integer:

int nn = RNG.int64();
Definition: RandomNumberGenerator.h:1231
uint64_t int64()
Draw a 64 bit random number.
Definition: RandomNumberGenerator.cc:119

Random double with Gaussian distribution:

double rr = RNG.doub();
double doub()
Draw a random number with double precision.
Definition: RandomNumberGenerator.cc:137
@ RNG_GAUSS
Gaussian deviates.
Definition: RandomNumberGenerator.h:1244
void setDistribution(const RNG_DISTR &distribution, const std::vector< double > &vec_params=std::vector< double >())
Set the distribution to draw random numbers from.
Definition: RandomNumberGenerator.cc:302

Purpose of this class

We offer two inherently 32 bit generators, and an inherently 64 bit generator (although all three need 64 bits space), as well as some convenience methods.

The goal is to have a generator suite that satisfies all needs for statistical filters / Monte Carlo methods (and not more), especially when working within MeteoIO. In a way, statistical filters are what ultimately justify this class, and therefore it is meant to be tailored to their needs (and be C++98).

So, if you are currently using this (cf. Appendix A):

srand( time(nullptr) );
return rand() % range;
return rand() / double(RAND_MAX + 1);

then switch to MeteoIO's RNG. If however you rely heavily on the best quality random numbers, maybe even crypto-secure, there are some links to dedicated libraries in the Bibliography. Apart from the generators and distributions, this class aims to take away all the small steps that are often quickly deemed good enough, i.e. generator choice, seeding, saving states, range calculations, ...

Note
There is an example program exercising most of the RNG's features in the /doc/examples folder.

Overview

What it can already do:

  • produce quality 64 bit, 32 bit and double random numbers with one simple call
  • doubles with different probability distributions
  • some probability density functions and cumulative distribution functions
  • make use of quality hardware and time seeds
  • fast downscaling of random numbers to a range
  • true floating point random numbers without rounding
  • can be resumed from a saved state
  • sidesteps some widespread misuse of quick & dirty solutions
  • sidesteps some issues with the insidious standard library
  • offers a ready-to-use interface for implementing new distributions (or even generators)
  • passes statistical tests
  • good benchmarks for the generator cores

What's left to do:

  • some distributions
  • Monte Carlo sampling template for arbitrary distribution functions

Random integers

To draw random integers, you can use either the int32() or int64() function call to receive 32 or 64 bit pseudo-random values, respectively:

uint32_t rn = RNG.int32();
uint64_t rm = RNG.int64();
uint32_t int32()
Draw a 32 bit random number.
Definition: RandomNumberGenerator.cc:128
Note
The generators guarantee exactly 32 or 64 random bits, so the appropriate types defined by inttypes.h are suited best. However, if you can live with conversion compiler warnings then any integer type will do. Of course, you can also simply use the type uintNN_t is mapped to on your machine. In short, int rn = RNG.int32() will work for a quick try.

Double values

Usually, you will simply get a double value like this:

double rr = RNG.doub();

You can call the doub() function with an RNG_BOUND argument including or excluding 0 and 1 (cf. the enum below). This can only be done for the uniform distribution, where it's clear what the borders are.

double rr = RNG.doub(RNG_AEXCBINC); //make sure it's not 0
rr = log(rr);
@ RNG_AEXCBINC
(0, 1]
Definition: RandomNumberGenerator.h:1256

Uniform random double values are quite hard to generate. The code example at Ref. [TC14] provides a method to do it, which is to interpret a random stream of bits as fractional part of the binary expansion of a number in [0, 1]. The file also goes into details about why other methods are troublesome if we rely on quality, e. g. sensitive random searches on a plane due to the gap size of 1/2^(bits).

In short, the doub() function returns a double within [0, 1] that is rounded to the nearest 1/2^64th. To get around this, you can set true_double to use an algorithm that calculates doubles in [0, 1] without the usual limitation of floating point randoms being on a grid (but then you must use RNG_AINCBINC and guard that in your own code if it must not happen even once).

double rr;
do {
rr = RNG.doub(RNG_AINCBINC, true); //get a random float on continuous axis
} while (rr == 0.); //make sure it's not 0
rr = log(rr);
@ RNG_AINCBINC
[0, 1]
Definition: RandomNumberGenerator.h:1254

Distributions / Random deviates

For doubles, you can select from a number of distribution functions.

For example, you can draw a Student-t variate like this:

double rr = RNG.doub();
@ RNG_STUDENTT
Student-t deviates.
Definition: RandomNumberGenerator.h:1248

If you don't set any distribution parameters, they will be defaulted (cf. Distribution parameters).

So far, the following deviates are available, defined by their probability density:

  • Uniform, RNG_UNIFORM:

    \[ f(x) = \frac 1{b-a} \quad \mathrm{for} \quad a \le x \le b, \]

  • Gauss (= Normal), RNG_GAUSS:

    \[ f(x \mid\mu,\sigma^2)=\frac{1}{\sqrt{2\pi\sigma^2}}\exp\left(-\frac{(x-\mu)^2}{2\sigma^2}\right) \quad \mathrm{for} \quad -\infty < x < +\infty \]

  • Gamma, RNG_GAMMA:

    \[ f(x)=\frac{\beta^\alpha}{\Gamma(\alpha)}x^{\alpha-1}e^{-\beta x} \quad \mathrm{for} \quad x > 0 \]

    \[ \alpha > 0, \quad \beta > 0 \]

  • Student-t, RNG_STUDENTT:

    \[ f_\nu(x)=\frac{\Gamma\left(\frac{\nu+1}{2}\right)} {\sqrt{\nu\pi}~\Gamma\left(\frac{\nu}{2}\right)} \left(1+\frac{x^{2}}{\nu}\right)^{-\frac{\nu+1}{2}} \quad \mathrm{for} \quad -\infty < x < +\infty \]

    \[ \nu > 0 \]

  • Chi-squared, RNG_CHISQUARED:

    \[ Z\sim\mathcal{N}(0,1) \rightarrow Z^2\sim\chi^2(1) \rightarrow Y=\chi^2(r_1)+\ldots+\chi^2(r_\nu)\sim\chi^2(r_1+\ldots+r_\nu) \]

    \[ \nu > 0 \]

  • Beta, RNG_BETA:

    \[ f(x)=\frac{1}{\mathrm{B}(\alpha, \beta)} x^{\alpha-1}(1-x)^{\beta-1} \quad \mathrm{for} \quad 0 < x < 1 \]

    \[ \alpha > 0, \quad \beta > 0 \]

  • F, RNG_F:

    \[ f(x|\nu _1, \nu _2) = m^{\frac{\nu _1}{2}} \nu _2^{\frac{\nu _2}{2}} \cdot \frac{\Gamma (\frac{\nu _1}{2}+\frac{\nu _2}{2})}{\Gamma (\frac{\nu _1}{2}) \Gamma(\frac{\nu _2}{2})} \cdot \frac{x^{\frac{\nu _1}{2}-1}} {(\nu _1 x+\nu _2)^\frac{\nu _1+\nu _2}{2}} \quad \mathrm{for} \quad x \geq 0 \]

Note
Integer values are always drawn with Uniform distribution, so if you really need a different one, for now you'll have to draw a double value and multiply accordingly.

Distribution parameters

When you change the distribution, you switch to a completely new one. The distribution parameters have to be provided each time, or they will be defaulted.

You can set distribution parameters in two ways:

  1. By setting (getting) them one by one after a distribution has been set:
    RNG.setDistributionParameter("mean", 5.);
    RNG.setDistributionParameter("sigma", 2.);
    double mean_out = RNG.getDistributionParameter("mean");
    void setDistributionParameter(const std::string &param_name, const double &param_val)
    Set single distribution parameter.
    Definition: RandomNumberGenerator.cc:470
    double getDistributionParameter(const std::string &param_name) const
    Retrieve single distribution parameter.
    Definition: RandomNumberGenerator.cc:407
  2. Via accessing the DistributionParameters vector directly. A std::vector<double> is provided with input parameters, or it has the output stored to it. This vector is given to the setDistribution() or getDistribution() call (the latter also returns the distribution type).

    Set distribution and parameters:

    std::vector<double> distribution_params;
    const double alpha = 1.2, beta = 1.;
    distribution_params.push_back(alpha);
    distribution_params.push_back(beta);
    @ RNG_GAMMA
    Gamma deviates.
    Definition: RandomNumberGenerator.h:1246

    Get distribution and parameters:

    distribution_params.clear();
    const mio::RandomNumberGenerator::RNG_DISTR dist_t = RNG.getDistribution(distribution_params);
    const double alpha_out = distribution_params.at(0); //check doc for indices
    RNG_DISTR getDistribution(std::vector< double > &vec_params) const
    Set the state of the RNG (seed the RNG)
    Definition: RandomNumberGenerator.cc:287
    RNG_DISTR
    Definition: RandomNumberGenerator.h:1242

Here are the names within MeteoIO, arguments and default values of the distributions described above:

DistributionIndexParameterDescriptionDefault value
RNG_UNIFORM--no parameteres-
RNG_GAUSS = RNG_NORMAL1meancenter of curve0
2sigmastandard deviation1
RNG_GAMMA1alphashape parameter 11
2betashape parameter 21
RNG_CHISQUARED1nunumber of degrees of freedom1
RNG_STUDENTT1nunumber of degrees of freedom1
2meancenter of curve0
3sigmastandard deviation1
RNG_BETA1alphashape parameter 11
2betashape parameter 21
RNG_F1nu1degrees of freedom in numerator1
2nu2degrees of freedom in denominator1

PDFs and CDFs

So far, the probability density function and cumulative distribution function are available for the Gauss distribution like this:

const double rg = RNG.doub();
std::cout << "Drew: " << rg << std::endl;
std::cout << std::setprecision(4) << "Probability to hit a number close to this one: "
<< RNG.pdf(rg)*100 << " %" << std::endl;
std::cout << "Probability to hit below this number: " << RNG.cdf(rg)*100 << " %" << std::endl;
double cdf(const double &xx)
Cumulative distribution function of selected distribution (integrated distribution function)
Definition: RandomNumberGenerator.cc:196
double pdf(const double &xx)
Probability density function of selected distribution.
Definition: RandomNumberGenerator.cc:186

Range calculations

You can draw 32 and 64 bit integers in a given range like this:

uint32_t rn = RNG.range32(10, 20);
uint64_t rn = RNG.range64(100, 2000);
uint64_t range64(const uint64_t &aa, const uint64_t &bb)
64 bit random number in an interval [aa, bb]
Definition: RandomNumberGenerator.cc:207
uint32_t range32(const uint32_t &aa, const uint32_t &bb)
32 bit random number in an interval [aa, bb]
Definition: RandomNumberGenerator.cc:232

Note that whatever you do, for an arbitrary count of random numbers you cannot downscale them and keep the distribution completely intact (although "non-trivial" methods are under investigation) due to the Pigeonhole principle. The only way not to distort the (uniform) distribution is to generate lots of numbers and reject out of boundary values. This is done by the trueRange32() function with a default 1e6 tries before resorting to downscaling (indicated by the return boolean). You can crank this up, but to state the obvious if the range gets small this gets costly quickly.

uint32_t rt;
const bool true_range_success = RNG.trueRange32(100, 3000, rt);
bool trueRange32(const uint32_t &aa, const uint32_t &bb, uint32_t &result, const unsigned int &nmax=1e6)
Random integer in a range without distribution distortions.
Definition: RandomNumberGenerator.cc:248

Random number generator algorithms

Three algorithms are available, namely the Mersenne Twister, a "classical" combined generator, and a rather new algorithm with promising statistical qualities. You can set them when initializing the RNG like this:

@ RNG_XOR
Combined generator.
Definition: RandomNumberGenerator.h:1235
@ RNG_PCG
Permuted linear congruential generator.
Definition: RandomNumberGenerator.h:1236
@ RNG_MTW
Mersenne Twister generator.
Definition: RandomNumberGenerator.h:1237

Mersenne Twister

  • Implementation of the wide-spread Mersenne Twister algorithm by M. Matsumoto and T. Nishimura (Ref. [MN98]).
  • By using 624 internal states, the period is extremely long.
  • This does not make it crypto-secure (the state can be derived from 624 random numbers), but it passes many statistical tests and is the standard RNG in numerous well-known software packages.
  • It needs a few kB buffer size, which is relatively large compared to the other generators.
  • Facts: size: 32 bit, period: 2^19937-1 (Mersenne prime) ~ 4.3e6001

Combined generator

  • Generator with xor, shift and multiplication
  • This is a fast combined generator that should be suitable for all but very special Monte Carlo applications. Since more than one internal states are being propagated and combined to the output, this makes it somewhat less predictable than similar generators.
  • Seed with any value except vv.
  • Facts: size: 64 bit, period: ~3.138e57

PCG

  • Permuted linear congruential generator by Prof. Melissa O'Neill
  • Range is overestimated, and this generator performs very well in statistical tests, i. e. it is less predictable than related generators. Even smaller versions with only 32 bit entropy pass SmallCrunch, which is only barely theoretically possible. The key element is the hashing function from the internal states to the random number. The algorithm author describes this RNG family in her paper (Ref. [MO14]) and offers a huge sophisticated C-library for free download with from tiny to 128 bit generators.
  • You should seed true 64 bit values or discard the first numbers.
  • If drawing 64 bit naturally is slow on your machine, try this one.
  • Facts: size: 32 bit, period: ~2^64 ~ 1.8e19
Note
The numbers were subjected to the dieharder random number test suite (Ref. [RB03]), passing most tests (while rand() fails horribly). Here is a quick benchmark (cf. Appendix B for dieharder results):
Generator Bit # per second
XOR 64 5.73e7
XOR 32 7.04e7
PCG 32 13.57e7
MTW 32 5.19e7
MTW GAUSS 0.84e7
crand 32 17.21e7
STL MTW 32 7.69e7
STL GAUSS 0.48e7
Hardware 64 2.61e4

Tested on Intel Core i7-7700HQ CPU @ 2.80GHz with 32GB RAM.

Seeding

Each time a RNG is constructed, it auto-seeds from hardware noise, or if that fails by hashing the system time. If the system time is the same when you initialize more generators at once, it will still seed differently. Successful hardware noise can be checked with getHardwareSeedSuccess(), and it's also noted in the toString() info. Manually seeding the generator is done after the fact with setState(), for example, to resume experiments after the state was saved via getState(). Note that this will not reset if you seed the generator yourself; i. e. if you seed from hardware and then later resume the chain by re-seeding, it will still show as hardware seeded. Finally, we offer the getUniqueSeed() function, so if you have set up your calculations with a grandfathered in, better, faster, ... RNG we can at least help with the seeding.

Example: By default, the Mersenne Twister initializes its 624 states with a linear congruential generator and then mixes that together with 64 hardware noise (resp. time hash) values. If you wanted to seed all 624 internal states with hardware noise (or time hashes) you could do it like this:

std::vector<uint64_t> seed_array;
uint64_t seed;
bool success(false);
for (size_t i = 0; i < 624; ++i) {
success = MTW.getUniqueSeed(seed);
seed_array.push_back(seed);
}
MTW.setState(seed_array);

You can retrieve the generator's state to later resume experiments at exactly this point:

std::vector<uint64_t> out_seed;
RNG.getState(out_seed);
RN2.setState(out_seed);
void getState(std::vector< uint64_t > &ovec_seed) const
Get the state of the RNG to save for later continuation.
Definition: RandomNumberGenerator.cc:269
void setState(const std::vector< uint64_t > &ivec_seed)
Set the state of the RNG (seed the RNG)
Definition: RandomNumberGenerator.cc:278

Developer's guide

For developers of statistical filters it may be important to be able to implement custom probability distributions, for example for an empirical nonlinear sensor response. This class tries to be easy to expand in that regard. There are comment markers in the header and source files leading with "`CUSTOM_DIST step #`: ..." in the 7 places you need to register your custom distribution functions at. These 7 steps are:

  1. Give your distribution a name within MeteoIO
  2. Put your functions' prototypes in the header
  3. Point to your distribution function in the generic setDistribution() function, and use the interface to the caller to set your distribution parameters
  4. Give a small output info string
  5. Write your distribution function, its pdf and cdf (if only to throw a not-implemented error)
  6. If you want, you can map your parameters to names in the get- and setDistributionParameter() functions.
  7. Map a string shorthand to the name of your distribution.

Bibliography

  • [AS73] Abramowitz, Stegun. Handbook of Mathematical Functions. Applied Mathematics Series 55, 10th edition, 1973.
  • [DK81] Donald E. Knuth. The art of computer programming 2. Addison-Wesley series in computer science and information processing, 2nd edition, 1981.
  • [GM03] George Marsaglia. Xorshift RNGs. Journal of Statistical Software, Articles, 8/14, 2003.
  • [MN98] Makoto Matsumoto and Takuji Nishimura. Mersenne Twister: A 623-dimensionally equidistributed uniform pseudo-random number generator. ACM Transactions on Modeling and Computer Simulation, 8/1, 1998. http://www.math.sci.hiroshima-u.ac.jp/~m-mat/MT/emt.html
  • [MO14] Melissa O'Neill. PCG: A family of simple fast space-efficient statistically good algorithms for random number generation. Harvey Mudd College, 2014. http://www.pcg-random.org
  • [MT00] G. Marsaglia and W. Tsang. A simple method for generating gamma variables. ACM Transactions on Mathematical Software, 26/3, 2000.
  • [NR3] Press, Teukolsky, Vetterling, Flannery. Numerical Recipes. The Art of Scientific Computing. Cambridge University Press, 3rd edition, 2007.
  • [PE97] Pierre L'Ecuyer. Distribution properties of multiply-with-carry random number generators. Mathematics of Computation, 66/218i, 1997.
  • [PE99] Pierr L'Ecuyer. Tables of linear congruential generators of different sizes and good lattice structure. Mathematics of Computation, 68/225, 1999. Errata for the paper, read on 18-10-23)
  • [RB03] Robert G. Brown. Dieharder: A random number test suite. http://webhome.phy.duke.edu/~rgb/General/dieharder.php
  • [TC14] Taylor R. Campbell. http://mumble.net/~campbell/tmp/random_real.c (read on 18-10-23)

Appendix A

Why is

srand( time(nullptr) );
return rand() % range;
return rand() / double(RAND_MAX + 1);

bad?

  • A purely linear congruential RNG has purely bad statistical qualities
  • Quiet type collision between time() and srand()
  • (% range) distorts the distribution at the borders and (range + 1) should be used
  • Careful not to hit RAND_MAX = INT_MAX, maybe ((double)RAND_MAX) + 1.
Triplets of random numbers generated by a poor linear congruential generator.

Why is

std::random_device RNG;
std::seed_seq seed{RNG()};
std::mt19937 RNG_MT(seed);

not good?

  • A 624 state Mersenne Twister is seeded with a single 32 bit value, not a sequence
  • Leads to statistical flaws; some numbers are never drawn
  • std::seed_seq isn't a bijection like it's supposed to be (as of C++17)
  • Can produce zero-state

Appendix B

Random number quality summary: The RNG performs as expected and passes statistical tests within reason.

The generators were subjected to the test suite dieharder, alongside with the hardware device, rand(), and a state of the art crypto-generator. Please refer to the man page for this project for an interpretation of the results. In short:

  • The p-number denotes how likely it is that a perfect generator would produce this sequence.
  • Values below 0.05 and above 0.95 are usually considered bad. However, if a generator does not produce p-values below 0.05 in 5% of the tests, this is equally bad, since the p-value itself is a uniform test statistic.
  • This means that in a full test run, a handful of "weak" results are expected!

Even with default values, dieharder uses up massive amounts of random numbers, and it is designed to be able to push all generators to failure and make a stronger assessment than the ambiguous "weak". However, this also pushes the runtime. A billion (1e9) random numbers were used per test (ca. 10.7 GB) and still the file was rewound 1266 times per test; some "weak" results may be due to this. The hardware seed was piped to dieharder for a continuous flow of random words.

#=============================================================================#
# dieharder version 3.31.1 Copyright 2003 Robert G. Brown #
#=============================================================================#
# ---------- C's rand() function ----------
#=============================================================================#
test_name |ntup| tsamples |psamples| p-value |Assessment
#=============================================================================#
diehard_birthdays| 0| 100| 100|0.92354359| PASSED
diehard_operm5| 0| 1000000| 100|0.05552403| PASSED
diehard_rank_32x32| 0| 40000| 100|0.00000000| FAILED
diehard_rank_6x8| 0| 100000| 100|0.98180550| PASSED
diehard_bitstream| 0| 2097152| 100|0.00000000| FAILED
diehard_opso| 0| 2097152| 100|0.10873316| PASSED
diehard_oqso| 0| 2097152| 100|0.98545797| PASSED
diehard_dna| 0| 2097152| 100|0.00000000| FAILED
diehard_count_1s_str| 0| 256000| 100|0.00000000| FAILED
diehard_count_1s_byt| 0| 256000| 100|0.00000000| FAILED
diehard_parking_lot| 0| 12000| 100|0.00000000| FAILED
diehard_2dsphere| 2| 8000| 100|0.00000000| FAILED
diehard_3dsphere| 3| 4000| 100|0.00000000| FAILED
diehard_squeeze| 0| 100000| 100|0.00000000| FAILED
diehard_sums| 0| 100| 100|0.00000000| FAILED
diehard_runs| 0| 100000| 100|0.57088403| PASSED
diehard_runs| 0| 100000| 100|0.72115015| PASSED
diehard_craps| 0| 200000| 100|0.00000000| FAILED
diehard_craps| 0| 200000| 100|0.00000000| FAILED
marsaglia_tsang_gcd| 0| 10000000| 100|0.00000000| FAILED
marsaglia_tsang_gcd| 0| 10000000| 100|0.48796452| PASSED
sts_monobit| 1| 100000| 100|0.00000000| FAILED
sts_runs| 2| 100000| 100|0.00000000| FAILED
sts_serial| 1| 100000| 100|0.00000000| FAILED
sts_serial| 2| 100000| 100|0.00000000| FAILED
sts_serial| 3| 100000| 100|0.00000000| FAILED
sts_serial| 3| 100000| 100|0.89813018| PASSED
sts_serial| 4| 100000| 100|0.00000000| FAILED
sts_serial| 4| 100000| 100|0.75692532| PASSED
sts_serial| 5| 100000| 100|0.00000000| FAILED
sts_serial| 5| 100000| 100|0.69863769| PASSED
sts_serial| 6| 100000| 100|0.00000000| FAILED
sts_serial| 6| 100000| 100|0.97959672| PASSED
sts_serial| 7| 100000| 100|0.00000000| FAILED
sts_serial| 7| 100000| 100|0.35286943| PASSED
sts_serial| 8| 100000| 100|0.00000000| FAILED
sts_serial| 8| 100000| 100|0.02669475| PASSED
sts_serial| 9| 100000| 100|0.00000000| FAILED
sts_serial| 9| 100000| 100|0.10088919| PASSED
sts_serial| 10| 100000| 100|0.00000000| FAILED
sts_serial| 10| 100000| 100|0.67626624| PASSED
sts_serial| 11| 100000| 100|0.00000000| FAILED
sts_serial| 11| 100000| 100|0.24830326| PASSED
sts_serial| 12| 100000| 100|0.00000000| FAILED
sts_serial| 12| 100000| 100|0.51859521| PASSED
sts_serial| 13| 100000| 100|0.00000000| FAILED
sts_serial| 13| 100000| 100|0.54684078| PASSED
sts_serial| 14| 100000| 100|0.00000000| FAILED
sts_serial| 14| 100000| 100|0.24749310| PASSED
sts_serial| 15| 100000| 100|0.00000000| FAILED
sts_serial| 15| 100000| 100|0.28317200| PASSED
sts_serial| 16| 100000| 100|0.00000000| FAILED
sts_serial| 16| 100000| 100|0.96702381| PASSED
rgb_bitdist| 1| 100000| 100|0.00000000| FAILED
rgb_bitdist| 2| 100000| 100|0.00000000| FAILED
rgb_bitdist| 3| 100000| 100|0.00000000| FAILED
rgb_bitdist| 4| 100000| 100|0.00000000| FAILED
rgb_bitdist| 5| 100000| 100|0.00000000| FAILED
rgb_bitdist| 6| 100000| 100|0.00000000| FAILED
rgb_bitdist| 7| 100000| 100|0.00000000| FAILED
rgb_bitdist| 8| 100000| 100|0.00000000| FAILED
rgb_bitdist| 9| 100000| 100|0.00000000| FAILED
rgb_bitdist| 10| 100000| 100|0.00000000| FAILED
rgb_bitdist| 11| 100000| 100|0.00000000| FAILED
rgb_bitdist| 12| 100000| 100|0.00000000| FAILED
rgb_minimum_distance| 2| 10000| 1000|0.00000000| FAILED
rgb_minimum_distance| 3| 10000| 1000|0.00000000| FAILED
rgb_minimum_distance| 4| 10000| 1000|0.00000000| FAILED
rgb_minimum_distance| 5| 10000| 1000|0.00000000| FAILED
rgb_permutations| 2| 100000| 100|0.70561378| PASSED
rgb_permutations| 3| 100000| 100|0.78449684| PASSED
rgb_permutations| 4| 100000| 100|0.74532512| PASSED
rgb_permutations| 5| 100000| 100|0.64539281| PASSED
rgb_lagged_sum| 0| 1000000| 100|0.00000000| FAILED
rgb_lagged_sum| 1| 1000000| 100|0.00000000| FAILED
rgb_lagged_sum| 2| 1000000| 100|0.00000000| FAILED
rgb_lagged_sum| 3| 1000000| 100|0.00000000| FAILED
rgb_lagged_sum| 4| 1000000| 100|0.00000000| FAILED
rgb_lagged_sum| 5| 1000000| 100|0.00000000| FAILED
rgb_lagged_sum| 6| 1000000| 100|0.00000000| FAILED
rgb_lagged_sum| 7| 1000000| 100|0.00000000| FAILED
rgb_lagged_sum| 8| 1000000| 100|0.00000000| FAILED
rgb_lagged_sum| 9| 1000000| 100|0.00000000| FAILED
rgb_lagged_sum| 10| 1000000| 100|0.00000000| FAILED
rgb_lagged_sum| 11| 1000000| 100|0.00000000| FAILED
rgb_lagged_sum| 12| 1000000| 100|0.00000000| FAILED
rgb_lagged_sum| 13| 1000000| 100|0.00000000| FAILED
rgb_lagged_sum| 14| 1000000| 100|0.00000000| FAILED
rgb_lagged_sum| 15| 1000000| 100|0.00000000| FAILED
rgb_lagged_sum| 16| 1000000| 100|0.00000000| FAILED
rgb_lagged_sum| 17| 1000000| 100|0.00000000| FAILED
rgb_lagged_sum| 18| 1000000| 100|0.00000000| FAILED
rgb_lagged_sum| 19| 1000000| 100|0.00000000| FAILED
rgb_lagged_sum| 20| 1000000| 100|0.00000000| FAILED
rgb_lagged_sum| 21| 1000000| 100|0.00000000| FAILED
rgb_lagged_sum| 22| 1000000| 100|0.00000000| FAILED
rgb_lagged_sum| 23| 1000000| 100|0.00000000| FAILED
rgb_lagged_sum| 24| 1000000| 100|0.00000000| FAILED
rgb_lagged_sum| 25| 1000000| 100|0.00000000| FAILED
rgb_lagged_sum| 26| 1000000| 100|0.00000000| FAILED
rgb_lagged_sum| 27| 1000000| 100|0.00000000| FAILED
rgb_lagged_sum| 28| 1000000| 100|0.00000000| FAILED
rgb_lagged_sum| 29| 1000000| 100|0.00000000| FAILED
rgb_lagged_sum| 30| 1000000| 100|0.00000000| FAILED
rgb_lagged_sum| 31| 1000000| 100|0.00000000| FAILED
rgb_lagged_sum| 32| 1000000| 100|0.00000000| FAILED
rgb_kstest_test| 0| 10000| 1000|0.00000000| FAILED
dab_bytedistrib| 0| 51200000| 1|0.00000000| FAILED
dab_dct| 256| 50000| 1|0.00000000| FAILED
dab_filltree| 32| 15000000| 1|0.27021810| PASSED
dab_filltree| 32| 15000000| 1|0.19091321| PASSED
dab_filltree2| 0| 5000000| 1|0.00000000| FAILED
dab_filltree2| 1| 5000000| 1|0.00000000| FAILED
dab_monobit2| 12| 65000000| 1|1.00000000| FAILED
#=============================================================================#
# ---------- MeteoIO's XOR generator ----------
#=============================================================================#
test_name |ntup| tsamples |psamples| p-value |Assessment
#=============================================================================#
# The file file_input was rewound 88 times
diehard_birthdays| 0| 100| 100|0.12045448| PASSED
diehard_operm5| 0| 1000000| 100|0.68176209| PASSED
diehard_rank_32x32| 0| 40000| 100|0.38302444| PASSED
diehard_rank_6x8| 0| 100000| 100|0.70959824| PASSED
diehard_bitstream| 0| 2097152| 100|0.13085472| PASSED
diehard_opso| 0| 2097152| 100|0.54027846| PASSED
diehard_oqso| 0| 2097152| 100|0.96236177| PASSED
diehard_dna| 0| 2097152| 100|0.10289128| PASSED
diehard_count_1s_str| 0| 256000| 100|0.72671076| PASSED
diehard_count_1s_byt| 0| 256000| 100|0.74290626| PASSED
diehard_parking_lot| 0| 12000| 100|0.45602493| PASSED
diehard_2dsphere| 2| 8000| 100|0.60117959| PASSED
diehard_3dsphere| 3| 4000| 100|0.86642811| PASSED
diehard_squeeze| 0| 100000| 100|0.43870967| PASSED
diehard_sums| 0| 100| 100|0.64147029| PASSED
diehard_runs| 0| 100000| 100|0.77132323| PASSED
diehard_runs| 0| 100000| 100|0.20403616| PASSED
diehard_craps| 0| 200000| 100|0.93550853| PASSED
diehard_craps| 0| 200000| 100|0.18756198| PASSED
marsaglia_tsang_gcd| 0| 10000000| 100|0.81899786| PASSED
marsaglia_tsang_gcd| 0| 10000000| 100|0.36808636| PASSED
sts_monobit| 1| 100000| 100|0.21979851| PASSED
sts_runs| 2| 100000| 100|0.02284836| PASSED
sts_serial| 1| 100000| 100|0.82689663| PASSED
sts_serial| 2| 100000| 100|0.99713803| WEAK
sts_serial| 3| 100000| 100|0.18497580| PASSED
sts_serial| 3| 100000| 100|0.05111402| PASSED
sts_serial| 4| 100000| 100|0.07758415| PASSED
sts_serial| 4| 100000| 100|0.45390764| PASSED
sts_serial| 5| 100000| 100|0.12301209| PASSED
sts_serial| 5| 100000| 100|0.92184175| PASSED
sts_serial| 6| 100000| 100|0.86072819| PASSED
sts_serial| 6| 100000| 100|0.86762303| PASSED
sts_serial| 7| 100000| 100|0.51949446| PASSED
sts_serial| 7| 100000| 100|0.94327933| PASSED
sts_serial| 8| 100000| 100|0.03901476| PASSED
sts_serial| 8| 100000| 100|0.14766057| PASSED
sts_serial| 9| 100000| 100|0.32167782| PASSED
sts_serial| 9| 100000| 100|0.85589005| PASSED
sts_serial| 10| 100000| 100|0.27561237| PASSED
sts_serial| 10| 100000| 100|0.84549731| PASSED
sts_serial| 11| 100000| 100|0.81071716| PASSED
sts_serial| 11| 100000| 100|0.19921550| PASSED
sts_serial| 12| 100000| 100|0.77836792| PASSED
sts_serial| 12| 100000| 100|0.94431528| PASSED
sts_serial| 13| 100000| 100|0.57153659| PASSED
sts_serial| 13| 100000| 100|0.15777123| PASSED
sts_serial| 14| 100000| 100|0.70324315| PASSED
sts_serial| 14| 100000| 100|0.69293364| PASSED
sts_serial| 15| 100000| 100|0.13673480| PASSED
sts_serial| 15| 100000| 100|0.22964883| PASSED
sts_serial| 16| 100000| 100|0.61626290| PASSED
sts_serial| 16| 100000| 100|0.82155600| PASSED
rgb_bitdist| 1| 100000| 100|0.58867260| PASSED
rgb_bitdist| 2| 100000| 100|0.24041805| PASSED
rgb_bitdist| 3| 100000| 100|0.90631007| PASSED
rgb_bitdist| 4| 100000| 100|0.27943794| PASSED
rgb_bitdist| 5| 100000| 100|0.30483125| PASSED
rgb_bitdist| 6| 100000| 100|0.91824913| PASSED
rgb_bitdist| 7| 100000| 100|0.06336972| PASSED
rgb_bitdist| 8| 100000| 100|0.43249369| PASSED
rgb_bitdist| 9| 100000| 100|0.36140764| PASSED
rgb_bitdist| 10| 100000| 100|0.67846826| PASSED
rgb_bitdist| 11| 100000| 100|0.88433105| PASSED
rgb_bitdist| 12| 100000| 100|0.93743658| PASSED
rgb_minimum_distance| 2| 10000| 1000|0.57731305| PASSED
rgb_minimum_distance| 3| 10000| 1000|0.37804172| PASSED
rgb_minimum_distance| 4| 10000| 1000|0.07249321| PASSED
rgb_minimum_distance| 5| 10000| 1000|0.21384664| PASSED
rgb_permutations| 2| 100000| 100|0.71271095| PASSED
rgb_permutations| 3| 100000| 100|0.38717961| PASSED
rgb_permutations| 4| 100000| 100|0.42087970| PASSED
rgb_permutations| 5| 100000| 100|0.63534024| PASSED
rgb_lagged_sum| 0| 1000000| 100|0.29051338| PASSED
rgb_lagged_sum| 1| 1000000| 100|0.70594074| PASSED
rgb_lagged_sum| 2| 1000000| 100|0.07553691| PASSED
rgb_lagged_sum| 3| 1000000| 100|0.96291303| PASSED
rgb_lagged_sum| 4| 1000000| 100|0.99029236| PASSED
rgb_lagged_sum| 5| 1000000| 100|0.58679539| PASSED
rgb_lagged_sum| 6| 1000000| 100|0.50294005| PASSED
rgb_lagged_sum| 7| 1000000| 100|0.51399132| PASSED
rgb_lagged_sum| 8| 1000000| 100|0.75970188| PASSED
rgb_lagged_sum| 9| 1000000| 100|0.64410621| PASSED
rgb_lagged_sum| 10| 1000000| 100|0.31410172| PASSED
rgb_lagged_sum| 11| 1000000| 100|0.49745204| PASSED
rgb_lagged_sum| 12| 1000000| 100|0.63601533| PASSED
rgb_lagged_sum| 13| 1000000| 100|0.67325992| PASSED
rgb_lagged_sum| 14| 1000000| 100|0.89565303| PASSED
rgb_lagged_sum| 15| 1000000| 100|0.22455715| PASSED
rgb_lagged_sum| 16| 1000000| 100|0.67474052| PASSED
rgb_lagged_sum| 17| 1000000| 100|0.86231270| PASSED
rgb_lagged_sum| 18| 1000000| 100|0.83858353| PASSED
rgb_lagged_sum| 19| 1000000| 100|0.76827640| PASSED
rgb_lagged_sum| 20| 1000000| 100|0.86217123| PASSED
rgb_lagged_sum| 21| 1000000| 100|0.82616486| PASSED
rgb_lagged_sum| 22| 1000000| 100|0.71637003| PASSED
rgb_lagged_sum| 23| 1000000| 100|0.10813332| PASSED
rgb_lagged_sum| 24| 1000000| 100|0.24239024| PASSED
rgb_lagged_sum| 25| 1000000| 100|0.49073378| PASSED
rgb_lagged_sum| 26| 1000000| 100|0.48516249| PASSED
rgb_lagged_sum| 27| 1000000| 100|0.50884127| PASSED
rgb_lagged_sum| 28| 1000000| 100|0.81183837| PASSED
rgb_lagged_sum| 29| 1000000| 100|0.51041034| PASSED
rgb_lagged_sum| 30| 1000000| 100|0.48703712| PASSED
rgb_lagged_sum| 31| 1000000| 100|0.37454423| PASSED
rgb_lagged_sum| 32| 1000000| 100|0.24603530| PASSED
rgb_kstest_test| 0| 10000| 1000|0.42451693| PASSED
dab_bytedistrib| 0| 51200000| 1|0.53431819| PASSED
dab_dct| 256| 50000| 1|0.49869025| PASSED
dab_filltree| 32| 15000000| 1|0.97577160| PASSED
dab_filltree| 32| 15000000| 1|0.78867864| PASSED
dab_filltree2| 0| 5000000| 1|0.92150183| PASSED
dab_filltree2| 1| 5000000| 1|0.86433669| PASSED
dab_monobit2| 12| 65000000| 1|0.68684982| PASSED
#=============================================================================#
# ---------- MeteoIO's PCG generator ----------
#=============================================================================#
test_name |ntup| tsamples |psamples| p-value |Assessment
#=============================================================================#
diehard_birthdays| 0| 100| 100|0.30611733| PASSED
diehard_operm5| 0| 1000000| 100|0.85337458| PASSED
diehard_rank_32x32| 0| 40000| 100|0.13861533| PASSED
diehard_rank_6x8| 0| 100000| 100|0.52227522| PASSED
diehard_bitstream| 0| 2097152| 100|0.97554697| PASSED
diehard_opso| 0| 2097152| 100|0.51552460| PASSED
diehard_oqso| 0| 2097152| 100|0.53264545| PASSED
diehard_dna| 0| 2097152| 100|0.61817549| PASSED
diehard_count_1s_str| 0| 256000| 100|0.99764411| WEAK
diehard_count_1s_byt| 0| 256000| 100|0.70976036| PASSED
diehard_parking_lot| 0| 12000| 100|0.87630056| PASSED
diehard_2dsphere| 2| 8000| 100|0.65075574| PASSED
diehard_3dsphere| 3| 4000| 100|0.05268042| PASSED
diehard_squeeze| 0| 100000| 100|0.62515552| PASSED
diehard_sums| 0| 100| 100|0.41680388| PASSED
diehard_runs| 0| 100000| 100|0.95126389| PASSED
diehard_runs| 0| 100000| 100|0.19295069| PASSED
diehard_craps| 0| 200000| 100|0.56451449| PASSED
diehard_craps| 0| 200000| 100|0.70173861| PASSED
marsaglia_tsang_gcd| 0| 10000000| 100|0.30987088| PASSED
marsaglia_tsang_gcd| 0| 10000000| 100|0.01578185| PASSED
sts_monobit| 1| 100000| 100|0.37553060| PASSED
sts_runs| 2| 100000| 100|0.58229832| PASSED
sts_serial| 1| 100000| 100|0.14004258| PASSED
sts_serial| 2| 100000| 100|0.11515591| PASSED
sts_serial| 3| 100000| 100|0.10588840| PASSED
sts_serial| 3| 100000| 100|0.31060254| PASSED
sts_serial| 4| 100000| 100|0.17520685| PASSED
sts_serial| 4| 100000| 100|0.88866431| PASSED
sts_serial| 5| 100000| 100|0.66233000| PASSED
sts_serial| 5| 100000| 100|0.84998032| PASSED
sts_serial| 6| 100000| 100|0.97556496| PASSED
sts_serial| 6| 100000| 100|0.32807318| PASSED
sts_serial| 7| 100000| 100|0.33865125| PASSED
sts_serial| 7| 100000| 100|0.26793941| PASSED
sts_serial| 8| 100000| 100|0.04701350| PASSED
sts_serial| 8| 100000| 100|0.34346444| PASSED
sts_serial| 9| 100000| 100|0.47954800| PASSED
sts_serial| 9| 100000| 100|0.58661247| PASSED
sts_serial| 10| 100000| 100|0.92972114| PASSED
sts_serial| 10| 100000| 100|0.39666160| PASSED
sts_serial| 11| 100000| 100|0.99744594| WEAK
sts_serial| 11| 100000| 100|0.55783437| PASSED
sts_serial| 12| 100000| 100|0.52260461| PASSED
sts_serial| 12| 100000| 100|0.72804698| PASSED
sts_serial| 13| 100000| 100|0.22331086| PASSED
sts_serial| 13| 100000| 100|0.99708143| WEAK
sts_serial| 14| 100000| 100|0.24692884| PASSED
sts_serial| 14| 100000| 100|0.11816038| PASSED
sts_serial| 15| 100000| 100|0.48867820| PASSED
sts_serial| 15| 100000| 100|0.64283400| PASSED
sts_serial| 16| 100000| 100|0.57175474| PASSED
sts_serial| 16| 100000| 100|0.80413247| PASSED
rgb_bitdist| 1| 100000| 100|0.78286393| PASSED
rgb_bitdist| 2| 100000| 100|0.41751882| PASSED
rgb_bitdist| 3| 100000| 100|0.99225317| PASSED
rgb_bitdist| 4| 100000| 100|0.26843160| PASSED
rgb_bitdist| 5| 100000| 100|0.41019859| PASSED
rgb_bitdist| 6| 100000| 100|0.00857013| PASSED
rgb_bitdist| 7| 100000| 100|0.35404539| PASSED
rgb_bitdist| 8| 100000| 100|0.78633455| PASSED
rgb_bitdist| 9| 100000| 100|0.94074517| PASSED
rgb_bitdist| 10| 100000| 100|0.98280710| PASSED
rgb_bitdist| 11| 100000| 100|0.36084540| PASSED
rgb_bitdist| 12| 100000| 100|0.10641731| PASSED
rgb_minimum_distance| 2| 10000| 1000|0.35560587| PASSED
rgb_minimum_distance| 3| 10000| 1000|0.78872602| PASSED
rgb_minimum_distance| 4| 10000| 1000|0.32824016| PASSED
rgb_minimum_distance| 5| 10000| 1000|0.62849158| PASSED
rgb_permutations| 2| 100000| 100|0.81115433| PASSED
rgb_permutations| 3| 100000| 100|0.95847665| PASSED
rgb_permutations| 4| 100000| 100|0.21432525| PASSED
rgb_permutations| 5| 100000| 100|0.32399489| PASSED
rgb_lagged_sum| 0| 1000000| 100|0.68964293| PASSED
rgb_lagged_sum| 1| 1000000| 100|0.99736916| WEAK
rgb_lagged_sum| 2| 1000000| 100|0.88914743| PASSED
rgb_lagged_sum| 3| 1000000| 100|0.69040691| PASSED
rgb_lagged_sum| 4| 1000000| 100|0.56425845| PASSED
rgb_lagged_sum| 5| 1000000| 100|0.72356996| PASSED
rgb_lagged_sum| 6| 1000000| 100|0.72350239| PASSED
rgb_lagged_sum| 7| 1000000| 100|0.96595205| PASSED
rgb_lagged_sum| 8| 1000000| 100|0.46167022| PASSED
rgb_lagged_sum| 9| 1000000| 100|0.66944952| PASSED
rgb_lagged_sum| 10| 1000000| 100|0.11552089| PASSED
rgb_lagged_sum| 11| 1000000| 100|0.47340674| PASSED
rgb_lagged_sum| 12| 1000000| 100|0.88912904| PASSED
rgb_lagged_sum| 13| 1000000| 100|0.85095112| PASSED
rgb_lagged_sum| 14| 1000000| 100|0.88590882| PASSED
rgb_lagged_sum| 15| 1000000| 100|0.96677181| PASSED
rgb_lagged_sum| 16| 1000000| 100|0.71908437| PASSED
rgb_lagged_sum| 17| 1000000| 100|0.37838020| PASSED
rgb_lagged_sum| 18| 1000000| 100|0.33524328| PASSED
rgb_lagged_sum| 19| 1000000| 100|0.05116258| PASSED
rgb_lagged_sum| 20| 1000000| 100|0.68284302| PASSED
rgb_lagged_sum| 21| 1000000| 100|0.18686823| PASSED
rgb_lagged_sum| 22| 1000000| 100|0.82848681| PASSED
rgb_lagged_sum| 23| 1000000| 100|0.68156360| PASSED
rgb_lagged_sum| 24| 1000000| 100|0.14155120| PASSED
rgb_lagged_sum| 25| 1000000| 100|0.85876186| PASSED
rgb_lagged_sum| 26| 1000000| 100|0.95982860| PASSED
rgb_lagged_sum| 27| 1000000| 100|0.93969005| PASSED
rgb_lagged_sum| 28| 1000000| 100|0.65020839| PASSED
rgb_lagged_sum| 29| 1000000| 100|0.88746765| PASSED
rgb_lagged_sum| 30| 1000000| 100|0.25119089| PASSED
rgb_lagged_sum| 31| 1000000| 100|0.29378236| PASSED
rgb_lagged_sum| 32| 1000000| 100|0.28869932| PASSED
rgb_kstest_test| 0| 10000| 1000|0.70588166| PASSED
dab_bytedistrib| 0| 51200000| 1|0.98774439| PASSED
dab_dct| 256| 50000| 1|0.19956807| PASSED
dab_filltree| 32| 15000000| 1|0.86399972| PASSED
dab_filltree| 32| 15000000| 1|0.10289422| PASSED
dab_filltree2| 0| 5000000| 1|0.05301030| PASSED
dab_filltree2| 1| 5000000| 1|0.97755469| PASSED
dab_monobit2| 12| 65000000| 1|0.49159043| PASSED
#=============================================================================#
# ---------- MeteoIO's Mersenne Twister ----------
#=============================================================================#
test_name |ntup| tsamples |psamples| p-value |Assessment
#=============================================================================#
diehard_birthdays| 0| 100| 100|0.82253247| PASSED
diehard_operm5| 0| 1000000| 100|0.21178957| PASSED
diehard_rank_32x32| 0| 40000| 100|0.31973686| PASSED
diehard_rank_6x8| 0| 100000| 100|0.89363426| PASSED
diehard_bitstream| 0| 2097152| 100|0.66751295| PASSED
diehard_opso| 0| 2097152| 100|0.68394748| PASSED
diehard_oqso| 0| 2097152| 100|0.97944473| PASSED
diehard_dna| 0| 2097152| 100|0.03763098| PASSED
diehard_count_1s_str| 0| 256000| 100|0.88917038| PASSED
diehard_count_1s_byt| 0| 256000| 100|0.13483835| PASSED
diehard_parking_lot| 0| 12000| 100|0.59549008| PASSED
diehard_2dsphere| 2| 8000| 100|0.79186313| PASSED
diehard_3dsphere| 3| 4000| 100|0.99908164| WEAK
diehard_squeeze| 0| 100000| 100|0.69206636| PASSED
diehard_sums| 0| 100| 100|0.00166453| WEAK
diehard_runs| 0| 100000| 100|0.10483337| PASSED
diehard_runs| 0| 100000| 100|0.14514648| PASSED
diehard_craps| 0| 200000| 100|0.81768910| PASSED
diehard_craps| 0| 200000| 100|0.20136565| PASSED
marsaglia_tsang_gcd| 0| 10000000| 100|0.10953859| PASSED
marsaglia_tsang_gcd| 0| 10000000| 100|0.13366752| PASSED
sts_monobit| 1| 100000| 100|0.91230913| PASSED
sts_runs| 2| 100000| 100|0.14335596| PASSED
sts_serial| 1| 100000| 100|0.49939858| PASSED
sts_serial| 2| 100000| 100|0.32024996| PASSED
sts_serial| 3| 100000| 100|0.90074110| PASSED
sts_serial| 3| 100000| 100|0.53246121| PASSED
sts_serial| 4| 100000| 100|0.08954348| PASSED
sts_serial| 4| 100000| 100|0.06305190| PASSED
sts_serial| 5| 100000| 100|0.01483603| PASSED
sts_serial| 5| 100000| 100|0.20630707| PASSED
sts_serial| 6| 100000| 100|0.42662161| PASSED
sts_serial| 6| 100000| 100|0.80660235| PASSED
sts_serial| 7| 100000| 100|0.00018564| WEAK
sts_serial| 7| 100000| 100|0.04110236| PASSED
sts_serial| 8| 100000| 100|0.20346520| PASSED
sts_serial| 8| 100000| 100|0.87953884| PASSED
sts_serial| 9| 100000| 100|0.43224325| PASSED
sts_serial| 9| 100000| 100|0.62598394| PASSED
sts_serial| 10| 100000| 100|0.80793422| PASSED
sts_serial| 10| 100000| 100|0.99492509| PASSED
sts_serial| 11| 100000| 100|0.91235510| PASSED
sts_serial| 11| 100000| 100|0.35551241| PASSED
sts_serial| 12| 100000| 100|0.27855050| PASSED
sts_serial| 12| 100000| 100|0.14287708| PASSED
sts_serial| 13| 100000| 100|0.47268781| PASSED
sts_serial| 13| 100000| 100|0.86584342| PASSED
sts_serial| 14| 100000| 100|0.79177664| PASSED
sts_serial| 14| 100000| 100|0.59808724| PASSED
sts_serial| 15| 100000| 100|0.07924232| PASSED
sts_serial| 15| 100000| 100|0.02515889| PASSED
sts_serial| 16| 100000| 100|0.09788033| PASSED
sts_serial| 16| 100000| 100|0.47327069| PASSED
rgb_bitdist| 1| 100000| 100|0.90859656| PASSED
rgb_bitdist| 2| 100000| 100|0.65219299| PASSED
rgb_bitdist| 3| 100000| 100|0.86767388| PASSED
rgb_bitdist| 4| 100000| 100|0.17870586| PASSED
rgb_bitdist| 5| 100000| 100|0.98686633| PASSED
rgb_bitdist| 6| 100000| 100|0.98019975| PASSED
rgb_bitdist| 7| 100000| 100|0.43739173| PASSED
rgb_bitdist| 8| 100000| 100|0.67351096| PASSED
rgb_bitdist| 9| 100000| 100|0.98913128| PASSED
rgb_bitdist| 10| 100000| 100|0.19281365| PASSED
rgb_bitdist| 11| 100000| 100|0.66186242| PASSED
rgb_bitdist| 12| 100000| 100|0.93080409| PASSED
rgb_minimum_distance| 2| 10000| 1000|0.40935556| PASSED
rgb_minimum_distance| 3| 10000| 1000|0.30879540| PASSED
rgb_minimum_distance| 4| 10000| 1000|0.59912826| PASSED
rgb_minimum_distance| 5| 10000| 1000|0.64492200| PASSED
rgb_permutations| 2| 100000| 100|0.90616743| PASSED
rgb_permutations| 3| 100000| 100|0.89734728| PASSED
rgb_permutations| 4| 100000| 100|0.57694248| PASSED
rgb_permutations| 5| 100000| 100|0.91845418| PASSED
rgb_lagged_sum| 0| 1000000| 100|0.51985981| PASSED
rgb_lagged_sum| 1| 1000000| 100|0.74730732| PASSED
rgb_lagged_sum| 2| 1000000| 100|0.96077254| PASSED
rgb_lagged_sum| 3| 1000000| 100|0.64992886| PASSED
rgb_lagged_sum| 4| 1000000| 100|0.08889463| PASSED
rgb_lagged_sum| 5| 1000000| 100|0.19288526| PASSED
rgb_lagged_sum| 6| 1000000| 100|0.30230622| PASSED
rgb_lagged_sum| 7| 1000000| 100|0.57923993| PASSED
rgb_lagged_sum| 8| 1000000| 100|0.96730234| PASSED
rgb_lagged_sum| 9| 1000000| 100|0.83008767| PASSED
rgb_lagged_sum| 10| 1000000| 100|0.62428422| PASSED
rgb_lagged_sum| 11| 1000000| 100|0.49734948| PASSED
rgb_lagged_sum| 12| 1000000| 100|0.81433084| PASSED
rgb_lagged_sum| 13| 1000000| 100|0.33109879| PASSED
rgb_lagged_sum| 14| 1000000| 100|0.84036199| PASSED
rgb_lagged_sum| 15| 1000000| 100|0.38518051| PASSED
rgb_lagged_sum| 16| 1000000| 100|0.67239247| PASSED
rgb_lagged_sum| 17| 1000000| 100|0.57189104| PASSED
rgb_lagged_sum| 18| 1000000| 100|0.11060300| PASSED
rgb_lagged_sum| 19| 1000000| 100|0.08207939| PASSED
rgb_lagged_sum| 20| 1000000| 100|0.96797696| PASSED
rgb_lagged_sum| 21| 1000000| 100|0.85276133| PASSED
rgb_lagged_sum| 22| 1000000| 100|0.62396333| PASSED
rgb_lagged_sum| 23| 1000000| 100|0.10741997| PASSED
rgb_lagged_sum| 24| 1000000| 100|0.00133815| WEAK
rgb_lagged_sum| 25| 1000000| 100|0.93440744| PASSED
rgb_lagged_sum| 26| 1000000| 100|0.62208635| PASSED
rgb_lagged_sum| 27| 1000000| 100|0.67716950| PASSED
rgb_lagged_sum| 28| 1000000| 100|0.84200928| PASSED
rgb_lagged_sum| 29| 1000000| 100|0.12315098| PASSED
rgb_lagged_sum| 30| 1000000| 100|0.85076946| PASSED
rgb_lagged_sum| 31| 1000000| 100|0.98674289| PASSED
rgb_lagged_sum| 32| 1000000| 100|0.98132089| PASSED
rgb_kstest_test| 0| 10000| 1000|0.85518280| PASSED
dab_bytedistrib| 0| 51200000| 1|0.03865242| PASSED
dab_dct| 256| 50000| 1|0.45151852| PASSED
dab_filltree| 32| 15000000| 1|0.43711867| PASSED
dab_filltree| 32| 15000000| 1|0.17002662| PASSED
dab_filltree2| 0| 5000000| 1|0.75275151| PASSED
dab_filltree2| 1| 5000000| 1|0.43152216| PASSED
dab_monobit2| 12| 65000000| 1|0.56354274| PASSED
#=============================================================================#
# ---------- MeteoIO's hardware seed through /dev/urandom ----------
#=============================================================================#
test_name |ntup| tsamples |psamples| p-value |Assessment
#=============================================================================#
diehard_birthdays| 0| 100| 100|0.34995998| PASSED
diehard_operm5| 0| 1000000| 100|0.32746493| PASSED
diehard_rank_32x32| 0| 40000| 100|0.45529134| PASSED
diehard_rank_6x8| 0| 100000| 100|0.45464472| PASSED
diehard_bitstream| 0| 2097152| 100|0.74174566| PASSED
diehard_opso| 0| 2097152| 100|0.85391211| PASSED
diehard_oqso| 0| 2097152| 100|0.03289769| PASSED
diehard_dna| 0| 2097152| 100|0.18584203| PASSED
diehard_count_1s_str| 0| 256000| 100|0.26587361| PASSED
diehard_count_1s_byt| 0| 256000| 100|0.76487483| PASSED
diehard_parking_lot| 0| 12000| 100|0.78148856| PASSED
diehard_2dsphere| 2| 8000| 100|0.00404096| WEAK
diehard_3dsphere| 3| 4000| 100|0.07650764| PASSED
diehard_squeeze| 0| 100000| 100|0.76720468| PASSED
diehard_sums| 0| 100| 100|0.15021486| PASSED
diehard_runs| 0| 100000| 100|0.73190429| PASSED
diehard_runs| 0| 100000| 100|0.11641745| PASSED
diehard_craps| 0| 200000| 100|0.51327580| PASSED
diehard_craps| 0| 200000| 100|0.91790555| PASSED
marsaglia_tsang_gcd| 0| 10000000| 100|0.95940428| PASSED
marsaglia_tsang_gcd| 0| 10000000| 100|0.71308014| PASSED
sts_monobit| 1| 100000| 100|0.95078372| PASSED
sts_runs| 2| 100000| 100|0.68447953| PASSED
sts_serial| 1| 100000| 100|0.94111724| PASSED
sts_serial| 2| 100000| 100|0.91997054| PASSED
sts_serial| 3| 100000| 100|0.33160131| PASSED
sts_serial| 3| 100000| 100|0.36015219| PASSED
sts_serial| 4| 100000| 100|0.35524447| PASSED
sts_serial| 4| 100000| 100|0.80131461| PASSED
sts_serial| 5| 100000| 100|0.40432154| PASSED
sts_serial| 5| 100000| 100|0.29670898| PASSED
sts_serial| 6| 100000| 100|0.19624684| PASSED
sts_serial| 6| 100000| 100|0.78008913| PASSED
sts_serial| 7| 100000| 100|0.48943065| PASSED
sts_serial| 7| 100000| 100|0.79865985| PASSED
sts_serial| 8| 100000| 100|0.10168430| PASSED
sts_serial| 8| 100000| 100|0.02619230| PASSED
sts_serial| 9| 100000| 100|0.47818037| PASSED
sts_serial| 9| 100000| 100|0.65931186| PASSED
sts_serial| 10| 100000| 100|0.57902903| PASSED
sts_serial| 10| 100000| 100|0.20841366| PASSED
sts_serial| 11| 100000| 100|0.20262420| PASSED
sts_serial| 11| 100000| 100|0.04916154| PASSED
sts_serial| 12| 100000| 100|0.36846324| PASSED
sts_serial| 12| 100000| 100|0.67908792| PASSED
sts_serial| 13| 100000| 100|0.28850671| PASSED
sts_serial| 13| 100000| 100|0.27326200| PASSED
sts_serial| 14| 100000| 100|0.18634033| PASSED
sts_serial| 14| 100000| 100|0.77790083| PASSED
sts_serial| 15| 100000| 100|0.07263144| PASSED
sts_serial| 15| 100000| 100|0.03880330| PASSED
sts_serial| 16| 100000| 100|0.15115141| PASSED
sts_serial| 16| 100000| 100|0.40485798| PASSED
rgb_bitdist| 1| 100000| 100|0.26387368| PASSED
rgb_bitdist| 2| 100000| 100|0.41977272| PASSED
rgb_bitdist| 3| 100000| 100|0.81246062| PASSED
rgb_bitdist| 4| 100000| 100|0.53180325| PASSED
rgb_bitdist| 5| 100000| 100|0.57916879| PASSED
rgb_bitdist| 6| 100000| 100|0.70438684| PASSED
rgb_bitdist| 7| 100000| 100|0.28560217| PASSED
rgb_bitdist| 8| 100000| 100|0.58521528| PASSED
rgb_bitdist| 9| 100000| 100|0.32013386| PASSED
rgb_bitdist| 10| 100000| 100|0.61939902| PASSED
rgb_bitdist| 11| 100000| 100|0.08846552| PASSED
rgb_bitdist| 12| 100000| 100|0.70170402| PASSED
rgb_minimum_distance| 2| 10000| 1000|0.02671540| PASSED
rgb_minimum_distance| 3| 10000| 1000|0.94613230| PASSED
rgb_minimum_distance| 4| 10000| 1000|0.54465261| PASSED
rgb_minimum_distance| 5| 10000| 1000|0.05533111| PASSED
rgb_permutations| 2| 100000| 100|0.24551831| PASSED
rgb_permutations| 3| 100000| 100|0.48396629| PASSED
rgb_permutations| 4| 100000| 100|0.39585636| PASSED
rgb_permutations| 5| 100000| 100|0.83858400| PASSED
rgb_lagged_sum| 0| 1000000| 100|0.52754001| PASSED
rgb_lagged_sum| 1| 1000000| 100|0.61577700| PASSED
rgb_lagged_sum| 2| 1000000| 100|0.97312394| PASSED
rgb_lagged_sum| 3| 1000000| 100|0.95265125| PASSED
rgb_lagged_sum| 4| 1000000| 100|0.61691352| PASSED
rgb_lagged_sum| 5| 1000000| 100|0.22625956| PASSED
rgb_lagged_sum| 6| 1000000| 100|0.44895384| PASSED
rgb_lagged_sum| 7| 1000000| 100|0.67063631| PASSED
rgb_lagged_sum| 8| 1000000| 100|0.80618162| PASSED
rgb_lagged_sum| 9| 1000000| 100|0.95412586| PASSED
rgb_lagged_sum| 10| 1000000| 100|0.43904741| PASSED
rgb_lagged_sum| 11| 1000000| 100|0.00027503| WEAK
rgb_lagged_sum| 12| 1000000| 100|0.59191040| PASSED
rgb_lagged_sum| 13| 1000000| 100|0.96138321| PASSED
rgb_lagged_sum| 14| 1000000| 100|0.23925605| PASSED
rgb_lagged_sum| 15| 1000000| 100|0.95895455| PASSED
rgb_lagged_sum| 16| 1000000| 100|0.86479019| PASSED
rgb_lagged_sum| 17| 1000000| 100|0.99271063| PASSED
rgb_lagged_sum| 18| 1000000| 100|0.07340033| PASSED
rgb_lagged_sum| 19| 1000000| 100|0.30090484| PASSED
rgb_lagged_sum| 20| 1000000| 100|0.07296293| PASSED
rgb_lagged_sum| 21| 1000000| 100|0.07541724| PASSED
rgb_lagged_sum| 22| 1000000| 100|0.99744800| WEAK
rgb_lagged_sum| 23| 1000000| 100|0.11186717| PASSED
rgb_lagged_sum| 24| 1000000| 100|0.99086821| PASSED
rgb_lagged_sum| 25| 1000000| 100|0.02194406| PASSED
rgb_lagged_sum| 26| 1000000| 100|0.92171159| PASSED
rgb_lagged_sum| 27| 1000000| 100|0.88602559| PASSED
rgb_lagged_sum| 28| 1000000| 100|0.98115042| PASSED
rgb_lagged_sum| 29| 1000000| 100|0.02165800| PASSED
rgb_lagged_sum| 30| 1000000| 100|0.61661470| PASSED
rgb_lagged_sum| 31| 1000000| 100|0.20261751| PASSED
rgb_lagged_sum| 32| 1000000| 100|0.59218537| PASSED
rgb_kstest_test| 0| 10000| 1000|0.27361658| PASSED
dab_bytedistrib| 0| 51200000| 1|0.94592564| PASSED
dab_dct| 256| 50000| 1|0.41859755| PASSED
dab_filltree| 32| 15000000| 1|0.76739133| PASSED
dab_filltree| 32| 15000000| 1|0.26581069| PASSED
dab_filltree2| 0| 5000000| 1|0.02605838| PASSED
dab_filltree2| 1| 5000000| 1|0.74389817| PASSED
dab_monobit2| 12| 65000000| 1|0.28647627| PASSED
#=============================================================================#
# ---------- Modern AES crypto generator ----------
rng_name |rands/second| Seed |
AES_OFB| 6.37e+06 |3232005462|
#=============================================================================#
test_name |ntup| tsamples |psamples| p-value |Assessment
#=============================================================================#
diehard_birthdays| 0| 100| 100|0.74435744| PASSED
diehard_operm5| 0| 1000000| 100|0.32904570| PASSED
diehard_rank_32x32| 0| 40000| 100|0.85406683| PASSED
diehard_rank_6x8| 0| 100000| 100|0.44255811| PASSED
diehard_bitstream| 0| 2097152| 100|0.51899516| PASSED
diehard_opso| 0| 2097152| 100|0.41191163| PASSED
diehard_oqso| 0| 2097152| 100|0.32705297| PASSED
diehard_dna| 0| 2097152| 100|0.00002584| WEAK
diehard_count_1s_str| 0| 256000| 100|0.26033835| PASSED
diehard_count_1s_byt| 0| 256000| 100|0.65992315| PASSED
diehard_parking_lot| 0| 12000| 100|0.49340705| PASSED
diehard_2dsphere| 2| 8000| 100|0.49486843| PASSED
diehard_3dsphere| 3| 4000| 100|0.69480866| PASSED
diehard_squeeze| 0| 100000| 100|0.24509533| PASSED
diehard_sums| 0| 100| 100|0.56234006| PASSED
diehard_runs| 0| 100000| 100|0.54652419| PASSED
diehard_runs| 0| 100000| 100|0.61171723| PASSED
diehard_craps| 0| 200000| 100|0.63350710| PASSED
diehard_craps| 0| 200000| 100|0.34587483| PASSED
marsaglia_tsang_gcd| 0| 10000000| 100|0.56466735| PASSED
marsaglia_tsang_gcd| 0| 10000000| 100|0.54209067| PASSED
sts_monobit| 1| 100000| 100|0.58300255| PASSED
sts_runs| 2| 100000| 100|0.07772198| PASSED
sts_serial| 1| 100000| 100|0.46471954| PASSED
sts_serial| 2| 100000| 100|0.15862944| PASSED
sts_serial| 3| 100000| 100|0.42568332| PASSED
sts_serial| 3| 100000| 100|0.68293665| PASSED
sts_serial| 4| 100000| 100|0.73346962| PASSED
sts_serial| 4| 100000| 100|0.61288736| PASSED
sts_serial| 5| 100000| 100|0.83897171| PASSED
sts_serial| 5| 100000| 100|0.84743739| PASSED
sts_serial| 6| 100000| 100|0.83328092| PASSED
sts_serial| 6| 100000| 100|0.72778824| PASSED
sts_serial| 7| 100000| 100|0.46201158| PASSED
sts_serial| 7| 100000| 100|0.61236377| PASSED
sts_serial| 8| 100000| 100|0.82259071| PASSED
sts_serial| 8| 100000| 100|0.14753897| PASSED
sts_serial| 9| 100000| 100|0.90776543| PASSED
sts_serial| 9| 100000| 100|0.70647508| PASSED
sts_serial| 10| 100000| 100|0.89681973| PASSED
sts_serial| 10| 100000| 100|0.46076365| PASSED
sts_serial| 11| 100000| 100|0.67367369| PASSED
sts_serial| 11| 100000| 100|0.37293255| PASSED
sts_serial| 12| 100000| 100|0.84656977| PASSED
sts_serial| 12| 100000| 100|0.76373116| PASSED
sts_serial| 13| 100000| 100|0.39318660| PASSED
sts_serial| 13| 100000| 100|0.80848926| PASSED
sts_serial| 14| 100000| 100|0.73276243| PASSED
sts_serial| 14| 100000| 100|0.89562062| PASSED
sts_serial| 15| 100000| 100|0.21683312| PASSED
sts_serial| 15| 100000| 100|0.50060895| PASSED
sts_serial| 16| 100000| 100|0.25979182| PASSED
sts_serial| 16| 100000| 100|0.81651079| PASSED
rgb_bitdist| 1| 100000| 100|0.35818457| PASSED
rgb_bitdist| 2| 100000| 100|0.18499831| PASSED
rgb_bitdist| 3| 100000| 100|0.83785965| PASSED
rgb_bitdist| 4| 100000| 100|0.13635981| PASSED
rgb_bitdist| 5| 100000| 100|0.80288280| PASSED
rgb_bitdist| 6| 100000| 100|0.26835958| PASSED
rgb_bitdist| 7| 100000| 100|0.65062732| PASSED
rgb_bitdist| 8| 100000| 100|0.93339630| PASSED
rgb_bitdist| 9| 100000| 100|0.65592490| PASSED
rgb_bitdist| 10| 100000| 100|0.92999418| PASSED
rgb_bitdist| 11| 100000| 100|0.08819807| PASSED
rgb_bitdist| 12| 100000| 100|0.83994867| PASSED
rgb_minimum_distance| 2| 10000| 1000|0.20870831| PASSED
rgb_minimum_distance| 3| 10000| 1000|0.14801298| PASSED
rgb_minimum_distance| 4| 10000| 1000|0.67682632| PASSED
rgb_minimum_distance| 5| 10000| 1000|0.52034430| PASSED
rgb_permutations| 2| 100000| 100|0.98884473| PASSED
rgb_permutations| 3| 100000| 100|0.14452545| PASSED
rgb_permutations| 4| 100000| 100|0.55317692| PASSED
rgb_permutations| 5| 100000| 100|0.36027454| PASSED
rgb_lagged_sum| 0| 1000000| 100|0.33982549| PASSED
rgb_lagged_sum| 1| 1000000| 100|0.31595934| PASSED
rgb_lagged_sum| 2| 1000000| 100|0.65722479| PASSED
rgb_lagged_sum| 3| 1000000| 100|0.99972394| WEAK
rgb_lagged_sum| 4| 1000000| 100|0.24595444| PASSED
rgb_lagged_sum| 5| 1000000| 100|0.51646049| PASSED
rgb_lagged_sum| 6| 1000000| 100|0.28270631| PASSED
rgb_lagged_sum| 7| 1000000| 100|0.58433720| PASSED
rgb_lagged_sum| 8| 1000000| 100|0.82820708| PASSED
rgb_lagged_sum| 9| 1000000| 100|0.74532620| PASSED
rgb_lagged_sum| 10| 1000000| 100|0.16422313| PASSED
rgb_lagged_sum| 11| 1000000| 100|0.43606954| PASSED
rgb_lagged_sum| 12| 1000000| 100|0.79274604| PASSED
rgb_lagged_sum| 13| 1000000| 100|0.57575293| PASSED
rgb_lagged_sum| 14| 1000000| 100|0.68237030| PASSED
rgb_lagged_sum| 15| 1000000| 100|0.89142299| PASSED
rgb_lagged_sum| 16| 1000000| 100|0.17680959| PASSED
rgb_lagged_sum| 17| 1000000| 100|0.46577456| PASSED
rgb_lagged_sum| 18| 1000000| 100|0.91604868| PASSED
rgb_lagged_sum| 19| 1000000| 100|0.89800891| PASSED
rgb_lagged_sum| 20| 1000000| 100|0.79179949| PASSED
rgb_lagged_sum| 21| 1000000| 100|0.52070680| PASSED
rgb_lagged_sum| 22| 1000000| 100|0.98784262| PASSED
rgb_lagged_sum| 23| 1000000| 100|0.59760181| PASSED
rgb_lagged_sum| 24| 1000000| 100|0.95193486| PASSED
rgb_lagged_sum| 25| 1000000| 100|0.55947446| PASSED
rgb_lagged_sum| 26| 1000000| 100|0.70940243| PASSED
rgb_lagged_sum| 27| 1000000| 100|0.13067759| PASSED
rgb_lagged_sum| 28| 1000000| 100|0.67353164| PASSED
rgb_lagged_sum| 29| 1000000| 100|0.39176071| PASSED
rgb_lagged_sum| 30| 1000000| 100|0.26042688| PASSED
rgb_lagged_sum| 31| 1000000| 100|0.94158603| PASSED
rgb_lagged_sum| 32| 1000000| 100|0.09088013| PASSED
rgb_kstest_test| 0| 10000| 1000|0.31359471| PASSED
dab_bytedistrib| 0| 51200000| 1|0.34836976| PASSED
dab_dct| 256| 50000| 1|0.58392168| PASSED
dab_filltree| 32| 15000000| 1|0.45835465| PASSED
dab_filltree| 32| 15000000| 1|0.54421849| PASSED
dab_filltree2| 0| 5000000| 1|0.16036442| PASSED
dab_filltree2| 1| 5000000| 1|0.77754723| PASSED
dab_monobit2| 12| 65000000| 1|0.66929755| PASSED

#include <RandomNumberGenerator.h>

Public Types

enum  RNG_TYPE { RNG_XOR , RNG_PCG , RNG_MTW }
 
enum  RNG_DISTR {
  RNG_UNIFORM , RNG_GAUSS , RNG_NORMAL , RNG_GAMMA ,
  RNG_CHISQUARED , RNG_STUDENTT , RNG_BETA , RNG_F
}
 
enum  RNG_BOUND { RNG_AINCBINC , RNG_AINCBEXC , RNG_AEXCBINC , RNG_AEXCBEXC }
 

Public Member Functions

 RandomNumberGenerator (const RNG_TYPE &type=RNG_XOR, const RNG_DISTR &distribution=RNG_UNIFORM, const std::vector< double > &distribution_params=std::vector< double >())
 Default constructor. More...
 
 RandomNumberGenerator (const RandomNumberGenerator &rng)
 Copy-constructor. More...
 
virtual ~RandomNumberGenerator ()
 Default destructor. Makes sure all states are freed from memory. More...
 
RandomNumberGeneratoroperator= (const RandomNumberGenerator &rng)
 Copy-operator. More...
 
uint64_t int64 ()
 Draw a 64 bit random number. More...
 
uint32_t int32 ()
 Draw a 32 bit random number. More...
 
double doub ()
 Draw a random number with double precision. More...
 
double doub (const RNG_BOUND &bounds, const bool &true_double=false)
 Draw a tuned double random number. More...
 
double draw ()
 Draw a random number with double precision. More...
 
double pdf (const double &xx)
 Probability density function of selected distribution. More...
 
double cdf (const double &xx)
 Cumulative distribution function of selected distribution (integrated distribution function) More...
 
uint64_t range64 (const uint64_t &aa, const uint64_t &bb)
 64 bit random number in an interval [aa, bb] More...
 
uint32_t range32 (const uint32_t &aa, const uint32_t &bb)
 32 bit random number in an interval [aa, bb] More...
 
bool trueRange32 (const uint32_t &aa, const uint32_t &bb, uint32_t &result, const unsigned int &nmax=1e6)
 Random integer in a range without distribution distortions. More...
 
void getState (std::vector< uint64_t > &ovec_seed) const
 Get the state of the RNG to save for later continuation. More...
 
void setState (const std::vector< uint64_t > &ivec_seed)
 Set the state of the RNG (seed the RNG) More...
 
RNG_DISTR getDistribution (std::vector< double > &vec_params) const
 Set the state of the RNG (seed the RNG) More...
 
void setDistribution (const RNG_DISTR &distribution, const std::vector< double > &vec_params=std::vector< double >())
 Set the distribution to draw random numbers from. More...
 
double getDistributionParameter (const std::string &param_name) const
 Retrieve single distribution parameter. More...
 
void setDistributionParameter (const std::string &param_name, const double &param_val)
 Set single distribution parameter. More...
 
bool getHardwareSeedSuccess () const
 Check if hardware noise could be read. More...
 
bool getUniqueSeed (uint64_t &store) const
 Get a proper 64 bit seeding value for the generator. More...
 
std::string toString ()
 Print some info about the selected generator. More...
 

Static Public Member Functions

static RNG_TYPE strToRngtype (const std::string &str)
 Get an RNG_TYPE from a string. More...
 
static RNG_DISTR strToRngdistr (const std::string &str)
 Get an RNG_DISTR from a string. More...
 

Member Enumeration Documentation

◆ RNG_BOUND

Enumerator
RNG_AINCBINC 

[0, 1]

RNG_AINCBEXC 

[0, 1)

RNG_AEXCBINC 

(0, 1]

RNG_AEXCBEXC 

(0, 1)

◆ RNG_DISTR

Enumerator
RNG_UNIFORM 

Uniform deviates.

RNG_GAUSS 

Gaussian deviates.

RNG_NORMAL 

= RNG_GAUSS

RNG_GAMMA 

Gamma deviates.

RNG_CHISQUARED 

Chi-Squared deviates.

RNG_STUDENTT 

Student-t deviates.

RNG_BETA 

Beta deviates.

RNG_F 

Fisher deviates.

◆ RNG_TYPE

Enumerator
RNG_XOR 

Combined generator.

RNG_PCG 

Permuted linear congruential generator.

RNG_MTW 

Mersenne Twister generator.

Constructor & Destructor Documentation

◆ RandomNumberGenerator() [1/2]

mio::RandomNumberGenerator::RandomNumberGenerator ( const RNG_TYPE type = RNG_XOR,
const RNG_DISTR distribution = RNG_UNIFORM,
const std::vector< double > &  distribution_params = std::vector<double>() 
)

Default constructor.

Parameters
typeRandom number generator algorithm
distributionDistribution of double random numbers
distribution_paramsParameters to shape the distribution functions This builds an RNG object with defaults or if supplied the desired distribution properties

◆ RandomNumberGenerator() [2/2]

mio::RandomNumberGenerator::RandomNumberGenerator ( const RandomNumberGenerator source)

Copy-constructor.

Parameters
sourceRNG to copy from This builds an RNG object that is in an identical state as the given RNG

◆ ~RandomNumberGenerator()

mio::RandomNumberGenerator::~RandomNumberGenerator ( )
virtual

Default destructor. Makes sure all states are freed from memory.

Member Function Documentation

◆ cdf()

double mio::RandomNumberGenerator::cdf ( const double &  xx)

Cumulative distribution function of selected distribution (integrated distribution function)

Parameters
xxPoint to evaluate function at
Returns
Probability to hit a number smaller than or equal to xx

◆ doub() [1/2]

double mio::RandomNumberGenerator::doub ( )

Draw a random number with double precision.

Returns
Double random number with the set distribution (default: Uniform)

◆ doub() [2/2]

double mio::RandomNumberGenerator::doub ( const RNG_BOUND bounds,
const bool &  true_double = false 
)

Draw a tuned double random number.

Parameters
boundsChoose if the boundaries 0 and 1 are included or not
true_doubleSet to true to use an algorithm that directly produces doubles without rounding from an integer

◆ draw()

double mio::RandomNumberGenerator::draw ( )

Draw a random number with double precision.

Returns
Double random number with Uniform distribution

◆ getDistribution()

RandomNumberGenerator::RNG_DISTR mio::RandomNumberGenerator::getDistribution ( std::vector< double > &  vec_params) const

Set the state of the RNG (seed the RNG)

Parameters
[out]vec_paramsA vector containing the number of seeds the generator needs

◆ getDistributionParameter()

double mio::RandomNumberGenerator::getDistributionParameter ( const std::string &  param_name) const

Retrieve single distribution parameter.

Parameters
param_nameName of the parameter (see section Distribution parameters)
Returns
Current value of the distribution parameter

◆ getHardwareSeedSuccess()

bool mio::RandomNumberGenerator::getHardwareSeedSuccess ( ) const

Check if hardware noise could be read.

Returns
True if the generator was seeded from hardware, false if the system time was used

◆ getState()

void mio::RandomNumberGenerator::getState ( std::vector< uint64_t > &  ovec_seed) const
virtual

Get the state of the RNG to save for later continuation.

Parameters
[out]ovec_seedVector that the internal states are saved to

Implements mio::RngCore.

◆ getUniqueSeed()

bool mio::RandomNumberGenerator::getUniqueSeed ( uint64_t &  store) const

Get a proper 64 bit seeding value for the generator.

Parameters
[out]storeThe result is stored here
Returns
True if the generator was seeded from hardware, false if the system time was mixed to a pseudo-random seed.

◆ int32()

uint32_t mio::RandomNumberGenerator::int32 ( )
virtual

Draw a 32 bit random number.

Returns
32 bit random number

Implements mio::RngCore.

◆ int64()

uint64_t mio::RandomNumberGenerator::int64 ( )
virtual

Draw a 64 bit random number.

Returns
64 bit random number

Implements mio::RngCore.

◆ operator=()

RandomNumberGenerator & mio::RandomNumberGenerator::operator= ( const RandomNumberGenerator source)

Copy-operator.

Parameters
sourceRNG to copy from
Returns
RNG object that is in an identical state as the given RNG

◆ pdf()

double mio::RandomNumberGenerator::pdf ( const double &  xx)

Probability density function of selected distribution.

Parameters
xxPoint to evaluate function at
Returns
Probability to hit a number close to xx

◆ range32()

uint32_t mio::RandomNumberGenerator::range32 ( const uint32_t &  aa,
const uint32_t &  bb 
)

32 bit random number in an interval [aa, bb]

Parameters
aaLower bound (included)
bbUpper bound (included)
Returns
Random integer between aa and bb

◆ range64()

uint64_t mio::RandomNumberGenerator::range64 ( const uint64_t &  aa,
const uint64_t &  bb 
)

64 bit random number in an interval [aa, bb]

Parameters
aaLower bound (included)
bbUpper bound (included)
Returns
Random integer between aa and bb

◆ setDistribution()

void mio::RandomNumberGenerator::setDistribution ( const RNG_DISTR distribution,
const std::vector< double > &  vec_params = std::vector<double>() 
)

Set the distribution to draw random numbers from.

Parameters
distributionDesired distribution
vec_paramsDistribution parameters (mean, standard deviation, shape, ...)

◆ setDistributionParameter()

void mio::RandomNumberGenerator::setDistributionParameter ( const std::string &  param_name,
const double &  param_val 
)

Set single distribution parameter.

Parameters
param_nameName of the parameter (see section Distribution parameters) to set
param_valValue to set

◆ setState()

void mio::RandomNumberGenerator::setState ( const std::vector< uint64_t > &  ivec_seed)
virtual

Set the state of the RNG (seed the RNG)

Parameters
ivec_seedA vector containing the number of seeds the generator needs

Implements mio::RngCore.

◆ strToRngdistr()

RandomNumberGenerator::RNG_DISTR mio::RandomNumberGenerator::strToRngdistr ( const std::string &  str)
static

Get an RNG_DISTR from a string.

Returns
Corresponding enum variable

◆ strToRngtype()

RandomNumberGenerator::RNG_TYPE mio::RandomNumberGenerator::strToRngtype ( const std::string &  str)
static

Get an RNG_TYPE from a string.

Returns
Corresponding enum variable

◆ toString()

std::string mio::RandomNumberGenerator::toString ( )

Print some info about the selected generator.

Returns
A small info string

◆ trueRange32()

bool mio::RandomNumberGenerator::trueRange32 ( const uint32_t &  aa,
const uint32_t &  bb,
uint32_t &  result,
const unsigned int &  nmax = 1e6 
)

Random integer in a range without distribution distortions.

Parameters
aaLower bound (included)
bbUpper bound (included)
[out]resultStores the found random number
nmaxMaximum number of tries before resorting to downscaling
Returns
Success of bruteforce method

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