Pretražite po imenu i prezimenu autora, mentora, urednika, prevoditelja

Napredna pretraga

Pregled bibliografske jedinice broj: 1199060

Determining the Influence of Hardware on the Execution Times of Trained Machine Learning Models


Baressi Šegota, Sandi; Glučina, Matko; Štifanić, Daniel; Musulin, Jelena; Lorencin, Ivan; Anđelić, Nikola; Car, Zlatan
Determining the Influence of Hardware on the Execution Times of Trained Machine Learning Models // RI-STEM-2022
Rijeka, 2022. str. 88-90 (predavanje, međunarodna recenzija, cjeloviti rad (in extenso), znanstveni)


CROSBI ID: 1199060 Za ispravke kontaktirajte CROSBI podršku putem web obrasca

Naslov
Determining the Influence of Hardware on the Execution Times of Trained Machine Learning Models
(Determining the Influence of Hardware on the Execution Times of Trained Machine Learning Models)

Autori
Baressi Šegota, Sandi ; Glučina, Matko ; Štifanić, Daniel ; Musulin, Jelena ; Lorencin, Ivan ; Anđelić, Nikola ; Car, Zlatan

Vrsta, podvrsta i kategorija rada
Radovi u zbornicima skupova, cjeloviti rad (in extenso), znanstveni

Izvornik
RI-STEM-2022 / - Rijeka, 2022, 88-90

ISBN
978-953-8246-26-5

Skup
Ri-STEM-2022

Mjesto i datum
Rijeka, Hrvatska, 8-9.6.2022

Vrsta sudjelovanja
Predavanje

Vrsta recenzije
Međunarodna recenzija

Ključne riječi
rtificial intelligence, convolutional neural networks, execution timing, high performance computing, hybrid systems, machine learning, model inference

Sažetak
While many discussions and observations have been made regarding the execution times of machine learning (ML) model training, not many researchers have shown concern regarding the execution times of trained models when the model are applied for inference. In this paper, the researchers observe the execution times of a realistic hybrid system consisting of a YOLOv3 based detection model and two classification models based on VGG16 and VGG19 convolutional neural networks. The provided hybrid model is tested on five different hardware configurations single core CPU execution, eight thread CPU execution, 48 thread CPU execution, single GOU execution and five GPU execution. The goal is to compare the execution times and determine the user satisfaction and ease of use on standard consumer hardware, readily available to the end users of the project, in comparison to a high performance computing architectures. The results show that the best performing architecture is a single GPU, but even the slowest solution (single core CPU), does not show an extremely slow time when inference is applied.

Izvorni jezik
Engleski

Znanstvena područja
Elektrotehnika, Računarstvo



POVEZANOST RADA


Projekti:
EK-EFRR-KK.01.2.2.03.0004 - Centar kompetencija za pametne gradove (CEKOM) (Car, Zlatan; Slavić, Nataša; Vilke, Siniša, EK - KK.01.2.2.03) ( POIROT)
EK-KF-KK.01.1.1.01.0009-1 - Napredne metode i tehnologije u znanosti o podatcima i kooperativnim sustavima - IJ za znanost o podatcima (Lončarić, Sven, EK - KK.01.1.1.01) ( POIROT)

Ustanove:
Tehnički fakultet, Rijeka

Poveznice na cjeloviti tekst rada:

Pristup cjelovitom tekstu rada

Citiraj ovu publikaciju:

Baressi Šegota, Sandi; Glučina, Matko; Štifanić, Daniel; Musulin, Jelena; Lorencin, Ivan; Anđelić, Nikola; Car, Zlatan
Determining the Influence of Hardware on the Execution Times of Trained Machine Learning Models // RI-STEM-2022
Rijeka, 2022. str. 88-90 (predavanje, međunarodna recenzija, cjeloviti rad (in extenso), znanstveni)
Baressi Šegota, S., Glučina, M., Štifanić, D., Musulin, J., Lorencin, I., Anđelić, N. & Car, Z. (2022) Determining the Influence of Hardware on the Execution Times of Trained Machine Learning Models. U: RI-STEM-2022.
@article{article, author = {Baressi \v{S}egota, Sandi and Glu\v{c}ina, Matko and \v{S}tifani\'{c}, Daniel and Musulin, Jelena and Lorencin, Ivan and An\djeli\'{c}, Nikola and Car, Zlatan}, year = {2022}, pages = {88-90}, keywords = {rtificial intelligence, convolutional neural networks, execution timing, high performance computing, hybrid systems, machine learning, model inference}, isbn = {978-953-8246-26-5}, title = {Determining the Influence of Hardware on the Execution Times of Trained Machine Learning Models}, keyword = {rtificial intelligence, convolutional neural networks, execution timing, high performance computing, hybrid systems, machine learning, model inference}, publisherplace = {Rijeka, Hrvatska} }
@article{article, author = {Baressi \v{S}egota, Sandi and Glu\v{c}ina, Matko and \v{S}tifani\'{c}, Daniel and Musulin, Jelena and Lorencin, Ivan and An\djeli\'{c}, Nikola and Car, Zlatan}, year = {2022}, pages = {88-90}, keywords = {rtificial intelligence, convolutional neural networks, execution timing, high performance computing, hybrid systems, machine learning, model inference}, isbn = {978-953-8246-26-5}, title = {Determining the Influence of Hardware on the Execution Times of Trained Machine Learning Models}, keyword = {rtificial intelligence, convolutional neural networks, execution timing, high performance computing, hybrid systems, machine learning, model inference}, publisherplace = {Rijeka, Hrvatska} }




Contrast
Increase Font
Decrease Font
Dyslexic Font