PESOM

Result code in IS VaVaI

<a href="https://www.isvavai.cz/riv?ss=detail&h=RIV%2F61989100%3A27740%2F18%3A10240397" target="_blank" >RIV/61989100:27740/18:10240397 - isvavai.cz</a>

Result on the web

<a href="https://github.com/It4innovations/PESOM" target="_blank" >https://github.com/It4innovations/PESOM</a>

DOI - Digital Object Identifier

—

Result language

angličtina

Original language name

PESOM

Original language description

It was created a parallel version of neural network algorithm (Self-organizing map), which focuses on working with sparse data on a large-scale dimension. The main goal is to create a new neural network in a short time. Many optimizations have been implemented, for example, for working with sparse data, updating neurons, batch updates, storing data in memory, etc. This implementation is stored in the GitHub repository at: https://github.com/It4innovations/PESOM

Czech name

—

Czech description

—

Type

R - Software

CEP classification

—

OECD FORD branch

10201 - Computer sciences, information science, bioinformathics (hardware development to be 2.2, social aspect to be 5.8)

Project

<a href="/en/project/LQ1602" target="_blank" >LQ1602: IT4Innovations excellence in science</a><br>

Continuities

P - Projekt vyzkumu a vyvoje financovany z verejnych zdroju (s odkazem do CEP)

Publication year

2018

Confidentiality

S - Úplné a pravdivé údaje o projektu nepodléhají ochraně podle zvláštních právních předpisů

Internal product ID

022/17-12-2018_SW

Technical parameters

Aplikace je napsaná a optimalizovaná v jazyce C++. Pro paralelní výpočet se používá kombinace MPI a OpenMP – na jednom výpočetním uzlu se využívají pro paralelizaci vlákna a pro komunikaci mezi výpočetními uzly je použito MPI.

Economical parameters

Neměřitelné

Owner IČO

61989100

Owner name

VŠB - Technical University of Ostrava