Metadata-Version: 2.1
Name: fastPSO
Version: 0.0.2
Summary: Fast parallel PSO library for Python with support for CPU and GPU multithreading.
Home-page: https://github.com/pribalta/fastPSO
Author: Pablo Ribalta Lorenzo
Author-email: pribalta@ieee.org
License: LICENSE.txt
Platform: UNKNOWN
Classifier: Programming Language :: Python :: 3
Classifier: License :: OSI Approved :: MIT License
Classifier: Operating System :: OS Independent

# fastPSO

[![Build Status](https://travis-ci.org/pribalta/fastPSO.svg?branch=master)](https://travis-ci.org/pribalta/fastPSO)

Fast parallel Particle Swarm Optimization package for Python

__fastPSO__ is an open source software library for Particle Swarm Optimization built with two goals in mind:
  * Speed
  * Parallelism

Its flexible architecture enables you to define complex objective functions, and to perform optimization in a __serial__ or __parallel__ setting. In addition, it offers detailed insights on the optimization process, helping practitioners profile their results.

## Installation

__pip__ __package__

```
pip install fastpso
```

### Requirements

  * numpy

## Getting started

tbd

## License

__fastPSO__ is available under *MIT License*

If you plan on using this software for scientific purposes, please cite our work:
```
@inproceedings{lorenzo2017particle,
  title={Particle swarm optimization for hyper-parameter selection in deep neural networks},
  author={Lorenzo, Pablo Ribalta et al.},
  booktitle={Proceedings of the Genetic and Evolutionary Computation Conference},
  pages={481--488},
  year={2017},
  organization={ACM}
}
```
```
@inproceedings{lorenzo2017hyper,
  title={Hyper-parameter selection in deep neural networks using parallel particle swarm optimization},
  author={Lorenzo, Pablo Ribalta et al.},
  booktitle={Proceedings of the Genetic and Evolutionary Computation Conference Companion},
  pages={1864--1871},
  year={2017},
  organization={ACM}
}
```





