/node-gsl

GNU Scientific Library (GSL) for NodeJS

Primary LanguageC++

           _ 
          | |
  __ _ ___| |
 / _` / __| |
| (_| \__ \ |
 \__, |___/_|
  __/ |      
 |___/

This project provide a binding between the GNU Scientific Library (GSL) and NodeJS.

At the moment, the library is partially integrated.

Installing

Via npm:

$ npm install gsl

Via git (or downloaded tarball):

$ git clone http://github.com/wdavidw/node-gsl.git
$ node-waf configure && node-waf

Random API

The library takes two forms, functions and iterator objects. Both respect the same names with different conventions. For exemple, obtaining an uniform random name can be done as random.get() as well as (new Random).get(). If you wished to provide a seed, then you'll respectivelly call random.get(seed) and (new Random(seed)).get()

Using an iterator objects make sense when using seeds and when performance is a concern.

Seeds are always optional and must be provided as unsigned integers. Deviations, used by gaussian functions, are float.

  • gsl.Random([seed])
    Construct a new iterator, seel below for available random methods.

  • gsl.random.get([seed])
    gsl.Random.get()
    Returns a random integer. The minimum and maximum values depend on the algorithm used, but all integers in the range [min,max] are equally likely. The values of min and max can determined using the auxiliary functions random.min() and random.max().

  • gsl.random.min()
    gsl.Random.min()
    Returns the smallest value that random.get() can return.

  • gsl.random.max()
    gsl.Random.max()
    Returns the largest value that random.get() can return.

  • gsl.random.uniform([seed])
    gsl.Random.uniform()
    Returns a double precision floating point number uniformly distributed in the range [0,1). The range includes 0.0 but excludes 1.0.

  • gsl.random.gaussian([seed], deviation)
    gsl.Random.gaussian(deviation)
    Returns a Gaussian random float with mean zero given a standart deviation as a float.

  • gsl.random.gaussianZiggurat([seed], deviation)
    gsl.Random.gaussianZiggurat(deviation)
    Same as random.gaussian but using the alternative Marsaglia-Tsang ziggurat method.

  • gsl.random.gaussianRatioMethod([seed], deviation)
    gsl.Random.gaussianRatioMethod(deviation)
    Same as random.gaussian but using the alternative Kinderman-Monahan-Leva ratio method.

  • gsl.random.poisson([seed], mean)
    gsl.Random.poisson(mean)
    Returns a random integer from the Poisson distribution given a provided mean as a float.

Exemple

var gsl = require('gsl'),
	seed = 50,
	deviation = 0.5;

console.log( gsl.random.gaussian(deviation) );
console.log( gsl.random.gaussian(seed, deviation) );

var iterator = new gsl.Random(seed);
console.log( iterator.gaussian(deviation) );
console.log( iterator.gaussian(deviation) );

Resources

Statistics API

Data are expected to be arrays of float numbers. Means are float numbers.

  • gsl.statistics.mean(data)
    Returns the arithmetic mean of data.

  • gsl.statistics.variance(data, [mean])
    Returns the estimated, or sample, variance of data.

  • gsl.statistics.sd(data, [mean])
    Returns the standard deviation defined as the square root of the variance defined above.

  • gsl.statistics.tss(data, [mean])
    Return the total sum of squares (TSS) of data about the mean. If mean is not provided, it is computed the same way as above.

  • gsl.statistics.varianceWithFixedMean(data, mean)
    Computes an unbiased estimate of the variance of data when the population mean mean of the underlying distribution is known a priori.

  • gsl.statistics.sdWithFixedMean(data, mean)
    Calculates the standard deviation of data for a fixed population mean mean. The result is the square root of the corresponding variance function.

Running the tests

Tests are executed with expresso. To install it, simply issue npm install expresso.

To run the tests expresso

Contributors