|
|
Generating Code from Textual Description of Functionality
Kačur, Ján ; Ondřej, Karel (referee) ; Smrž, Pavel (advisor)
The aim of this thesis was to design and implement system for code generation from textual description of functionality. In total, 2 systems were implemented. One of them served its purpose as a control prototype, the second one was the main product of this thesis. I focused on using smaller non-pre-trained models. Both systems used Transformer type model as their cores. The second system, unlike the first, used syntactic decomposition of both code and textual descriptions. Data used in both systems originated from project CodeSearchNet. Targer programming language to generate was Python. The second system achieved better quantitative results than the first one, with accuracy of 85% versus 60%. The system managed to auto-complete correct code to finish the function definition, with bigger time delay. This thesis is almost exclusively dedicated to the second system.
|
|
|
Fingerprint Identity Preserving Generative Adversarial Networks
Kačur, Ján ; Juránek, Roman (referee) ; Špaňhel, Jakub (advisor)
Táto práca sa sústredí na generovanie latentných odtlačkov prstov za pomoci Generatívnych oponentných neurónových sietí. Hlavnou úlohou je generovanie viacerých verzií latentných odtlačkov z čistého odtlačku, s rovnakou identitou. Identitu a štýl odtlačku by malo byť možné osobitne meniť. Zvolený postup sa zakladá na modeli AugNet. Navrhnutý algoritmus generuje latentné odtlačky z čistých binarizovaných odtlačkov a náhodného vektora, reprezentujúceho skreslenie, resp. štýl. V generátore sú použité AdaIN bloky na spojenie štýlu so vstupným odtlačkom. Je testovaných viacero trénovacích algoritmov, z ktorých WGAN-GP dosahuje najlepšie výsledky. Jednotlivé modely sú porovnávané kombináciou metrík FID a Rank-1 accuracy pri porovnávaní generovaných obrázkov s originálnymi vstupnými binarizovanými odtlačkami. Najlepšie modely sú vybrané ako Pareto optimálne kombinácie týchto 2 metrík.
|
|
|
Generating Code from Textual Description of Functionality
Kačur, Ján ; Ondřej, Karel (referee) ; Smrž, Pavel (advisor)
The aim of this thesis was to design and implement system for code generation from textual description of functionality. In total, 2 systems were implemented. One of them served its purpose as a control prototype, the second one was the main product of this thesis. I focused on using smaller non-pre-trained models. Both systems used Transformer type model as their cores. The second system, unlike the first, used syntactic decomposition of both code and textual descriptions. Data used in both systems originated from project CodeSearchNet. Targer programming language to generate was Python. The second system achieved better quantitative results than the first one, with accuracy of 85% versus 60%. The system managed to auto-complete correct code to finish the function definition, with bigger time delay. This thesis is almost exclusively dedicated to the second system.
|
guest ::
