https://github.com/glentner/parallelmt

The ParallelMT class creates a family of MT19937 objects seeded independently allowing for simultaneous access and generation from within a parallel environment.
https://github.com/glentner/parallelmt

Last synced: 6 days ago
JSON representation

The ParallelMT class creates a family of MT19937 objects seeded independently allowing for simultaneous access and generation from within a parallel environment.

• Host: GitHub
• URL: https://github.com/glentner/parallelmt
• Owner: glentner
• License: gpl-3.0
• Created: 2014-11-14T18:40:30.000Z (about 10 years ago)
• Default Branch: master
• Last Pushed: 2015-06-28T20:57:03.000Z (over 9 years ago)
• Last Synced: 2024-05-01T15:27:01.751Z (7 months ago)
• Language: C++
• Homepage:
• Size: 160 KB
• Stars: 0
• Watchers: 3
• Forks: 0
• Open Issues: 0
• Metadata Files:
  • Readme: README.md
  • License: LICENSE

Awesome Lists containing this project

README

        
#ParallelMT

####A Parallel Pseudo Random Number Generator for Monte Carlo Applications.

The ParallelMT class is an extension to the MT19937 class, both contained within a single MT

namespace. MT19937 is adapted from the original 64-bit pseudo random number generator

(Nishimura and Matsumoto, 2004). Here we have a class constructed with either a single seed or

an array seed. Methods include RandomInteger and RandomReal. The ParallelMT class creates a

family of MT19937 objects seeded independently allowing for simultaneous access and generation

from within a parallel environment.

### Example usage:

```c++

#include 

#include 

#define threads 4

int main(){

    // std::vector of length two designates `limits`

    double init[] = {1, 6};

    std::vector limits(init, sizeof(init)/sizeof(double));

    // construct using default seed

    ParallelMT pmt1();

    // construct using specified seed

    ParallelMT pmt2(0x1234ULL);

    // generate a random real numbers

    for (int i = 0; i < 1e6; i++)

    

    ...

    return 0;

}

```