|
|
Improving Bots Playing Starcraft II Game in PySC2 Environment
Krušina, Jan ; Škoda, Petr (referee) ; Smrž, Pavel (advisor)
The aim of this thesis is to create an automated system for playing a real-time strategy game Starcraft II. Learning from replays via supervised learning and reinforcement learning techniques are used for improving bot's behavior. The proposed system should be capable of playing the whole game utilizing PySC2 framework for machine learning. Performance of the bot is evaluated against the built-in scripted AI in the game.
|
|
|
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.
|
|
|
Authorship Identification
Fabiánek, Ondřej ; Škoda, Petr (referee) ; Smrž, Pavel (advisor)
This bachelor's thesis deals with authorship identification based on knowledge of author's previous texts. The aim is to analyze existing methods of authorship attribution and create a system, which is capable of highly successful authorship identification. The system is based on a multivariate analysis and specializes at English books. Part of the solution is also a graphic user interface.
|
|
|
Information Technologies in Psychology
Ličko, Jozef ; Grézl, František (referee) ; Smrž, Pavel (advisor)
We focus on characteristic traits recognition of the autor from his written text. This thesis, in particular, deals with the implementaion of Kreitler psychosemantics method. The result of our work includes our own vocabulary, that is used to assign one of the parameters from the method. Implemented solution is successful when used on a set of words that was used as a source for the vocabulary construction.
|
|
|
Wikipedia Page Classification
Suchý, Ondřej ; Otrusina, Lubomír (referee) ; Smrž, Pavel (advisor)
The goal of this paper is to design and implement a system for selection of Wikipedia articles relevant to a given topic in order to reduce the amount of memory taken by its offline version. The solution of this problem was achieved with use of methods from information retrieval and theirs implementation using Elasticsearch search engine. The system tries to determine the area of user's interest by given keywords and make a selection of articles from that area. This is achieved by measuring of similarity of articles and adding all articles from frequent categories in the selection. The sizes of the output files for queries over Simple English Wikipedia are usually below 30 MB.
|
|
|
A Monitoring and Aggregation Device for Large Amounts of Industry Data
Held, Oliver ; Pánek, Richard (referee) ; Smrž, Pavel (advisor)
This bachelor thesis describes hardware and software development of a monitoring device for the collection of industrial big data. The device is based on the development BeagleBone Black kit together with vibration sensors, temperature sensors and strain gauges. The objective of the thesis is to design a device with respect to low production cost, simplicity and possible modularity. The I2C bus and embedded AD converter are used for the collection of the data from the sensors. The data is sent to remote server using UDP protocol. Furthermore, custom strain gauge module is designed and tested. The practical contribution of the thesis is to test the platform functionality, that includes collection, aggregation and transfer of the data.
|
|
|
Searching Semantically Annotated Texts
Grešová, Katarína ; Smrž, Pavel (referee) ; Dytrych, Jaroslav (advisor)
This thesis deals with the issue of semantic searching over indexes of big text data. The aim of this thesis is to design and implement a search engine with web user interface enabling dynamical configuration of access to indexes and editing annotations in the text. The thesis analyzes the current search engine solution and its shortcomings, which results in a specification of requirements for a search engine that is suitable for common use and fulfils the potential of all search engine related tools. The thesis also describes the design, implementation and testing of the resulting system, which also includes an extension in a form of global constraints, which increases the accuracy of the requested search result description.
|
|
|
Extending Presentation Features of Jupyter Notebook
Pastorek, Peter ; Dytrych, Jaroslav (referee) ; Smrž, Pavel (advisor)
This work deals with Jupyter Notebook environment. It extends its abilities for presenting programming, allows downloading of materials from html documents, improves the look of presentations, allows automated preparation of the document for the presentation, editing of RISE settings used for presentation, and editing conversion of the document into LaTeX.
|
|
|
Videocodec - Videosequence Compression
Bařina, David ; Motlíček, Petr (referee) ; Smrž, Pavel (advisor)
This thesis deals with modern methods of a lossy still image and video compression. Wavelet transformation and SPIHT algorithm also belong to these methods. In second half of this thesis, a videocodec is implemented based on acquired knowledge. This codec uses Daubechies wavelets to analyse an image. Afterwards there is a modified SPIHT algorithm applied on gained coefficients. A lot of effort was put in order to optimize this computation. It is possible to use the created codec in Video for Windows, DirectShow and FFmpeg multimedia frameworks. At the end of this thesis, commonly used codecs are compared with newly created one.
|
|
|
Czech-Slovak Statistical Machine Translation
Astaloš, Lukáš ; Kouřil, Jan (referee) ; Smrž, Pavel (advisor)
The aim of this thesis is to design, implement and evaluate the translation system capable of translating texts from Czech to Slovak language. It describes theoretical foundations of statistical machine translation and then the phase of design and development of system. The chosen strategy was to build phrase-based translation model using one large parallel corpus in combination with language model from local institute of linguistics. The statistical machine translation Moses was used to achieve this goal. The vocabulary translation of unknown words using stemming was proposed and tested. Precision of build system was evaluated with BLEU score and it achieved comparable results with other systems.
|
guest ::
RSS feed.