Class: GSLng::RNG Abstract

Inherits:

Object

• Object
• GSLng::RNG

Defined in:

lib/gslng/rng/rng.rb,
lib/gslng/rng/uniform.rb,
lib/gslng/rng/gaussian.rb

Overview

This class is abstract.

You should use any of the descendant classes which implement the #sample method.

Random Number Generator class

Direct Known Subclasses

Gaussian, Uniform

Defined Under Namespace

Classes: Gaussian, Uniform

Class Method Summary

• .release(ptr) ⇒ Object

Instance Method Summary

• #initialize(generator_type = nil) ⇒ RNG constructor

  Create a Generator of the given type.

• #initialize_copy ⇒ Object

Constructor Details

#initialize(generator_type = nil) ⇒ RNG

Create a Generator of the given type

Parameters:

• generator_type (Symbol) (defaults to: nil) —

  the algorithm to use (without the gsl_rng prefix). Default is :mt19937.

See Also:

• http://www.gnu.org/software/gsl/manual/html_node/Random-number-generator-algorithms.html

# File 'lib/gslng/rng/rng.rb', line 8

def initialize(generator_type = nil)
  @type = generator_type
  if (@type.nil?) then @type = :mt19937 end # update comment above if changed
  
  type = GSLng.backend.send(:"gsl_rng_#{@type}")
  @ptr = FFI::AutoPointer.new(GSLng.backend.gsl_rng_alloc(type), RNG.method(:release))
end

Class Method Details

.release(ptr) ⇒ Object

# File 'lib/gslng/rng/rng.rb', line 16

def self.release(ptr)
  GSLng.backend.gsl_rng_free(ptr)
end

Instance Method Details

#initialize_copy ⇒ Object

# File 'lib/gslng/rng/rng.rb', line 20

def initialize_copy
  type = GSLng.backend.send(:"gsl_rng_#{@type}")
  @ptr = FFI::AutoPointer.new(GSLng.backend.gsl_rng_alloc(type), RNG.method(:release))
end