Implementacija i vrednovanje radnog okvira za razvoj modela dubokog učenja

master's thesis

• Title (english): Implementation and evaluation of a deep learning framework
• Author: Martin Čolja
• Mentor: Jan Šnajder (mentor)
• Committee member: Jan Šnajder (predsjednik povjerenstva)
• Committee member: Marko Đurasević (član povjerenstva)
• Committee member: Alan Jović (član povjerenstva)
• Granter: University of Zagreb; Faculty of Electrical Engineering and Computing; Zagreb
• Defense date and country: 2022-07-12, Croatia
• Scientific / art field, discipline and subdiscipline: TECHNICAL SCIENCES; Computing

Abstract

Duboko učenje je uzbudljivo područje za razvoj znanosti i industrijsku primjenu. Implementacija modela dubokog učenja zahtijeva učinkovite i fleksibilne alate s mogućnošću spajanja različitih modula u jedan kompleksniji model. Ovaj diplomski rad objašnjava postupak i implementacijske odluke potrebne za izradu jednostavne verzije takvog alata pomoćprogramskog jezika Swift. Programski okvir sadrži tri poveziva modula dubokog učenja i podršku za automatsku diferencijaciju. U radu su također prikazani primjeri korištenja i usporedbe navedenog programskog okvira s postojećim rješenjima u Pythonu.

Abstract (english)

Deep learning is a quickly developing area of research and industry application. Implementing deep learning models requires efficient and flexible tools with the ability to combine different modules into a single model. This thesis explains the steps and design decisions necessary to create a simple version of such a tool using Swift programming language. The framework features three connectable deep learning modules and automatic differentiation. Furthermore, the thesis demonstrates usage examples and compares the framework with existing solutions in Python

• Keywords: Duboko učenje; Swift; automatska diferencijacija; Pyhton
• Keywords (english): Deep learning; Swift; automatic differentiation; Python
• Language: english
• URN:NBN: urn:nbn:hr:168:712351
• Study programme: Title: Computing; Study programme type: university; Study level: graduate; Academic / professional title: magistar/magistra inženjer/inženjerka računarstva (magistar/magistra inženjer/inženjerka računarstva)
• Type of resource: Text
• File origin: Born digital
• Access conditions: Closed access
• Created on: 2023-01-13 14:45:37