prikaz prve stranice dokumenta Web aplikacija za trening i evaluaciju modela dubokog učenja koristeći različite skupove podataka
  Download
PDF 5.49 MB
master's thesis
Web aplikacija za trening i evaluaciju modela dubokog učenja koristeći različite skupove podataka
Zagreb: University of Zagreb, Faculty of Mechanical Engineering and Naval Architecture, 2024. urn:nbn:hr:235:996895
Mirić, Viktor
University of Zagreb
Faculty of Mechanical Engineering and Naval Architecture

Institutional repository: Repository of Faculty of Mechanical Engineering and Naval Architecture University of Zagreb

Cite this document

Mirić, V. (2024). Web aplikacija za trening i evaluaciju modela dubokog učenja koristeći različite skupove podataka (Master's thesis). Zagreb: University of Zagreb, Faculty of Mechanical Engineering and Naval Architecture. Retrieved from https://urn.nsk.hr/urn:nbn:hr:235:996895

Mirić, Viktor. "Web aplikacija za trening i evaluaciju modela dubokog učenja koristeći različite skupove podataka." Master's thesis, University of Zagreb, Faculty of Mechanical Engineering and Naval Architecture, 2024. https://urn.nsk.hr/urn:nbn:hr:235:996895

Mirić, Viktor. "Web aplikacija za trening i evaluaciju modela dubokog učenja koristeći različite skupove podataka." Master's thesis, University of Zagreb, Faculty of Mechanical Engineering and Naval Architecture, 2024. https://urn.nsk.hr/urn:nbn:hr:235:996895

Mirić, V. (2024). 'Web aplikacija za trening i evaluaciju modela dubokog učenja koristeći različite skupove podataka', Master's thesis, University of Zagreb, Faculty of Mechanical Engineering and Naval Architecture, accessed 18 December 2024, https://urn.nsk.hr/urn:nbn:hr:235:996895

Mirić V. Web aplikacija za trening i evaluaciju modela dubokog učenja koristeći različite skupove podataka [Master's thesis]. Zagreb: University of Zagreb, Faculty of Mechanical Engineering and Naval Architecture; 2024 [cited 2024 December 18] Available at: https://urn.nsk.hr/urn:nbn:hr:235:996895

V. Mirić, "Web aplikacija za trening i evaluaciju modela dubokog učenja koristeći različite skupove podataka", Master's thesis, University of Zagreb, Faculty of Mechanical Engineering and Naval Architecture, Zagreb, 2024. Available at: https://urn.nsk.hr/urn:nbn:hr:235:996895

Metadata
TitleWeb aplikacija za trening i evaluaciju modela dubokog učenja koristeći različite skupove podataka
Title (english)A web application for training and evaluating deep learning models using different datasets
AuthorViktor Mirić
MentorTomislav Stipančić (mentor)
GranterUniversity of Zagreb
Faculty of Mechanical Engineering and Naval Architecture
Zagreb
Defense date and country2024-12-05, Croatia
Scientific / art field,
discipline and subdiscipline		TECHNICAL SCIENCES
Mechanical Engineering
Abstract
Ovaj rad istražuje mogućnosti olakšavanja korištenja umjetne inteligencije i dubokog učenja, tehnologije koje sve više nalaze primjenu u različitim područjima. Motivacija ovog rada je razviti alat koji će korisnicima pružiti intuitivno sučelje za razvoj modela dubokog učenja bez potrebnog tehničkog znanja u toj domeni. Kao rješenje, razvijena je web aplikacija koja korisnicima omogućuje jednostavnu izradu i prilagodbu modela dubokog učenja. Sustav je implementiran korištenjem modernih tehnologija. Korisničko sučelje razvijeno je u programskom okviru Next.js, dok je pozadinska logika izrađena uz pomoć programskog okvira Flask. Aplikacija koristi PyTorch za treniranje neuronskih mreža, omogućujući korisnicima prilagodbu arhitekture mreže te vizualizaciju rezultata i performansi modela. Testiranje i evaluacija aplikacije provedena je na poznatom skupu podataka za klasifikaciju cvijeta perunike. Model je pokazao visoku preciznost, postigavši točnost predikcije od 100% na tesnom skupu podataka.
Abstract (english)
This paper explores the possibilities of facilitating the use of artificial intelligence and deep learning, technologies that are increasingly being used in various fields. The motivation for this paper is to develop a tool that will provide users with intuitive interface for developing deep learning models without the need for technical knowledge in this domain. As a result, a web application was developed that allows users to easily create and customize deep learning models. The system is implemented using modern technologies. The user interface is developed in the Next.js framework, while the backend logic is built using the Flask framework. The application uses PyTorch to train neural networks, allowing users to customize the network architecture and visualize the results and performance of the model. The application testing and evaluation was conducted on a well-known dataset for iris flower classification. The model demonstrated high precision, achieving 100% prediction accuracy on a test dataset.
Keywords
Keywords (english)
Languagecroatian
URN:NBNurn:nbn:hr:235:996895
Study programmeTitle: Mechanical Engineering; specializations in: Design, Process and Energy Engineering, Production Engineering, Engineering Modelling and Computer Simulation, Marine Engineering, Industrial Engineering and Management, Materials Engineering, Mechanics and Robotics, Autonomous Systems, Mechatronics and Robotics
Course: Industrial Engineering and Management
Study programme type: university
Study level: graduate
Academic / professional title: sveučilišni/a magistar/magistra inženjer/inženjerka strojarstva (sveučilišni/a magistar/magistra inženjer/inženjerka strojarstva)
Type of resourceText
File originBorn digital
Access conditionsOpen access
Terms of useLicence In copyright icon
Created on2024-12-01 14:34:53