Implementation of Software for Machine Learning and Deep Learning
Filonov, Egor (2021)
2021
All rights reserved. This publication is copyrighted. You may download, display and print it for Your own personal use. Commercial use is prohibited.
Julkaisun pysyvä osoite on
https://urn.fi/URN:NBN:fi:amk-202104306512
https://urn.fi/URN:NBN:fi:amk-202104306512
Tiivistelmä
The thesis studied how to create software for machine learning and deep learning. It also examined what the main components are in this kind of program and how are they implemented.
The central concepts that appear throughout the study are machine learning and deep learning. These abstractions are discussed in more detail during the explana-tion of the implementation process. Mathematical concepts, such as matrices and optimization algorithms, were used during the research. Moreover, topics from computer science, such as computation graphs, were applied to provide a better understanding of the final algorithm.
The results showed that the implementation of the software for machine learning and deep learning relies on the "training" algorithm and may vary depending on the type of problem approached with the software.
The central concepts that appear throughout the study are machine learning and deep learning. These abstractions are discussed in more detail during the explana-tion of the implementation process. Mathematical concepts, such as matrices and optimization algorithms, were used during the research. Moreover, topics from computer science, such as computation graphs, were applied to provide a better understanding of the final algorithm.
The results showed that the implementation of the software for machine learning and deep learning relies on the "training" algorithm and may vary depending on the type of problem approached with the software.