Institutional Repository of Vilnius University
Ranka rašytų skaitmenų atpažinimas naudojant mašininio mokymosi algoritmus /
Abstract
dc:descriptionIn this bachelor’s thesis, machine learning algorithms are used to solve the classification task to recognize handwritten digits. The aim of the work is to find out which functions can be used to obtain the highest neural network accuracy. To achieve this goal, results are compared using optimization and activation functions. For training and testing, MNIST handwritten digit database was selected, which consists of 70000 entries. The database consists of two different sources: handwritten digits of high school students and office workers. After evaluating neural network model, the most accurate model was determined.
Degree
thesis:*- Grantor dc:publisher
- Institutional Repository of Vilnius University
- Year dc:date
- 2023
Author and committee
dc:creator, dc:contributor.*- Author dc:creator
-
- Ribikauskaitė, Akvilė,
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:ELABAETD192845622&prefLang=en_US