random-distributions-cli

Command-line tool to generate samples of various random distributions

Usage

$ random-distributions-cli -n 5 normal 0 1
0.2691767185
-1.1783043912
0.5623772742
0.1078446497
0.2829313503

$ target/release/random-distributions-cli -n 3 uniform -- -1 1
0.0218074136
0.4381694923
-0.7817481855

$ random-distributions-cli --help
Usage: random-distributions-cli [-p <precision>] [-C] [-S <seed>] [-b <binary-format>] [-n <num-samples>] <command> [<args>]

Command-line tool to generate samples of various random distributions. Note that more single-value distributions that are mentioned in https://docs.rs/statrs/0.15.0/statrs/distribution/index.html are easy to add to the tool.

Options:
  -p, --precision   number of digits after decimal to print
  -C, --cumulative  add value of each sample to buffer
  -S, --seed        use specified seed instead for PRNG
  -b, --binary-format
                    output as binary numbers of specified format instead of
                    text. Valid formats are f{32,64}{be,le}, {u,s}8,
                    {u,s}{16,32,64}{le,be}. Out of range values are
                    clamped to valid ranges
  -n, --num-samples number of sampels to generate, instead of an infinite stream
  --help            display usage information

Commands:
  uniform           Uniform distribution
  normal            Normal, Gaussian distribution
  cauchy            Cauchy, Lorentz distribution - fat-tailed and continuous
  triangular        Triangular distribution - continuous
  studentt          Student-T distribution
  stable            General case of stable continuous distribution, generated by
                    CMS method. Note that this one is not checked by a
                    mathenatician, so don't rely on it to match other
                    implementation I used one hack about scaling to make it
                    match Cauchy and Normal distributions, but I am not sure if
                    it is correct.
  empirical         Discrete distribution that just endlessly randomly selects
                    one of specified values
  categorical       Discrete distribution that generates values according to
                    specified probabilities

Individual distributions may provide additional internal --help messages.