Skip to content

Latest commit

 

History

History
69 lines (49 loc) · 1.83 KB

README.md

File metadata and controls

69 lines (49 loc) · 1.83 KB

ClusOpt

A Data stream clusterer and hyper parameter optimizer using microservices.

Infraestructure

Prerequisites

  • python >= 3.6
  • pip
  • boost-thread
  • docker [optional]

Installing

Install all dependecies described in requirements.txt using pip.

pip3 install -r requirements.txt

To use docker just use the dockerfile

docker build -t clusopt .
docker run -it -p 3523:3523 clusopt bash

Usage

cd to the src directory if not already

cd src

to run primary node

./primary.py [ALGORITHM] [PATH OR URL TO CSV FILE]

NOTE: CSV File must be , separated, \n line ended and must not have a header.

to run replica node

./replica [IP ADDRESS TO THE PRIMARY NODE]

use -h and [ALGORITHM] -h in both primary and replica for more help

Published Works

[1] G. O. MACEDO and J. A. SILVA and D. M. L. FILHO. (2020). UM MODELO DE INFRAESTRUTURA DE MICROSSERVIÇOS PARA ALGORITMOS DE AGRUPAMENTO DE FLUXOS CONTÍNUOS DE DADOS. SODEBRAS Journal, 15(178), 48-52. https://doi.org/10.29367%2Fissn.1809-3957.15.2020.178.48

Authors

  • Giuliano Oliveira de Macedo - Coder giuliano-oliveira
  • Jonathan Andrade Silva - Advisor
  • Dionisio Leite Machado Filho - Advisor

Acknowledgments

Thanks to:

  • Dionisio for all the the tips to solve all the networking/cloud/virtualization related problems.
  • Jonathan to all stream clustering/clustering in general /clustering evaluation information used to build this system.
  • Marcel R. Ackermann et al. for the StreamKM++ algorithm - link
  • The university of Waikato for the MOA framework - link