This final thesis addresses the approach of standards and software for calculation of pressure vessels loaded by external pressure and a design of own calculation software, including a demonstration on chosen geometry. The approaches of standards and software are solved in the form of research. The practical part describes the developed software, selected technologies for development and results of the demonstration. The demonstration was carried out on two geometries and was successful.

The diploma thesis is focused on creating a calculation software for designing heat exchanger. The software will not calculate one single heat exchanger but will recommend all heat exchangers which meet the requirements. Then, user can decide which heat exchanger is best suited for the specific application and make a comprehensive calculation. Part of the diploma thesis is to verify correctness of the software and make a parametric studies. Parametric studies will be focused on baffle cut, baffle spacing, tube pitch, and tube thickness and its effect on heat transfer and pressure drop.

The aim of this diploma thesis is to study and design experimental improvement of the convolutional neural network for disease detection. Another goal is to extend the classifier with a new type of detection. he new type of detection is damage fingerprint by pressure. The experimentally improved convolutional network is implemented by PyTorch. The network detects which part of the fingerprint is damaged and draws this part into the fingerprint. Synthetic fingerprints are used when training the net. Real fingerprints are added to the synthetic fingerprints.

In this project, an algorithm has been implemented in Python to estimate the parameters describing the production and decay of a spin 1/2 baryon - antibaryon pair. This decay can give clues about a fundamental asymmetry between matter and antimatter. A model-independent formalism developed by the Uppsala hadron physics group and previously implemented in C++, has been shown to be a promising tool in the search for physics beyond the Standard Model (SM) of particle physics. The program developed in this work provides a more user-friendly alternative, and is intended to motivate further use of the formalism through a more maintainable, customizable and readable implementation. The hope is that this will expedite future research in the area of charge parity (CP)-violation and eventually lead to answers to questions such as why the universe consists of matter. A Monte-Carlo integrator is used for normalization and a Python library for function minimization. The program returns an estimation of the physics parameters including error estimation. Tests of statistical properties of the estimator, such as consistency and bias, have been performed. To speed up the implementation, the Just-In-Time compiler Numba has been employed which resulted in a speed increase of a factor 400 compared to plain Python code.

This thesis describes how to build a solution for transmitting data from an           Electroencephalography (EEG) device to a server in real-time while guiding the user through a number of predefined exercises. This solution will be used by Spinal Cord Injury (SCI) patients suffering from neuropathic pain, in order to understand if it is possible to predict such pain from EEG. The collected data will help clinicians analyze the brain activity data from patients who can submit the data from their home. To accomplish this development task, an application was built that connects to a portable EEG device, gather brain activity data from patients, guides patients through a set of imaginary tasks and sends the data to a server. This project made use of a Software Development Kit (SDK) for the Python programming language and a web sockets server written in JavaScript. The application was tested both in terms of usability and end-to-end latency, showing high usability and low latency. The proposed solution will support a clinical trial in Spain with 40 SCI patients.

This master’s thesis deals with using Text mining as a method to predict tags of articles. It describes the iterative way of handling big data files, parsing the data, cleaning the data and scoring of terms in article using TF-IDF. It describes in detail the flow of program written in programming language Python 3.4.3. The result of processing more than 1 million articles from Wikipedia database is a dictionary of English terms. By using this dictionary one is capable of determining the most important terms from article in corpus of articles. Relevancy of consequent tags proves the method used in this case.

This thesis deals with design and implementation of an investment model, which applies methods of Post-modern portfolio theory. Particle swarm optimization (PSO) metaheuristic was used for portfolio optimization and the parameters were analyzed with several experiments. Johnsons SU distribution was used for estimation of future returns as it proved to be the best of analyzed distributions. The result is software application written in Python, which is tested for stability and performance of model in extreme situations.

This thesis deals with the modifications of the framework designed for DNS servers testing. Framework is developed by NIC.CZ association and is used primarily for testing the DNS server Knot DNS. The aim of this work are modifications of the framework that will allow simpler testing with this framework, such as: support for multiple implementations of DNS servers, parallel testing, components dummy server and box-in-the-middle, division into multiple components and overall modification of the existing framework. Introduction of thesis is dedicated to the authoritative DNS servers and to the foundations of testing. The remaining part of the thesis deals with the state of the existing framework and the state and testing of modified framework.

Tato práce se zabývá běžnými problémy časových politik, které jsou využívány v rámci procesu přihlašování uživatelů. Jsou rozebrána řešení v některých jiných současných systémech. Dále je čtenáři představen projekt pro správu identit FreeIPA, autor se zaměřuje hlavně na správu uživatelů a politiky pro jejich autorizaci. Je také představen projekt SSSD se zaměřením na jeho propojení se systémem FreeIPA. Po vytvoření návrhu řešení problému časových politik je tento návrh implementován do systémů FreeIPA a SSSD.

Tato diplomová práce popisuje návrh a implementaci nástroje pro automatickou kategorizaci webových stránek. Cílem nástroje je aby byl schopen se z ukázkových webových stránek naučit, jak každá kategorie vypadá. Poté by měl nástroj zvládnout přiřadit naučené kategorie k dříve nespatřeným webovým stránkám. Nástroj by měl podporovat více kategorií a jazyků. Pro vývoj nástroje byly použity pokročilé techniky strojového učení, detekce jazyků a dolování dat. Nástroj je založen na open source knihovnách a je napsán v jazyce Python 3.3.