Institutional Repository of Vilnius University
Lietuviškos poezijos generavimas naudojantis dirbtiniais neuroniniais tinklais /
Abstract
dc:descriptionThis bachelor thesis analyzes the architectures of artificial neural network models based on long-short term memory and gated recurrent unit networks. These variants of recurrent neural networks are suitable for generating sequences because, unlike ordinary recurrent neural networks, they do not have the problem of vanishing gradients. Therefore, model architectures that uses long-shot term memory and gated recurrent neural networks were used to solve the problem of generating Lithuanian poetry. This work presents the preparation of data sets and their processing. Selected artificial neural network models are trained using the created data sets. Evaluations of training and text generation results are done.
Degree
thesis:*- Grantor dc:publisher
- Institutional Repository of Vilnius University
- Year dc:date
- 2022
Author and committee
dc:creator, dc:contributor.*- Author dc:creator
-
- Šivickas, Kipras,
Rights
dc:rights- Statement dc:rights
-
- info:eu-repo/semantics/openAccess
- Language dc:language
- lit
Identifiers
dc:identifier.*- Repository record dc:identifier
- https://repository.vu.lt/VU:ELABAETD146237121&prefLang=en_US