|
|
An application for testing automata learning algorithms
Vyšohlíd, Jan ; Hoffmann, Petr (advisor) ; Krčah, Peter (referee)
In the present work we study methods for testing regular inference algorithms. First there are introduced some theoretical basics for finite state automata and regular inference. Next we present some finite state automata learning algorithms, their principles and used testing methods, which find out algorithms quality via testing resulting automata. Following text makes the training and testing data generating process clear, describes format for saving this data and for saving finite state automata and finally the algorithms testing run alone, too. The application implementing present testing methods is a part of this work as well. It saves result statistics in chosen format. In appendices we append user and programmer documentation for this application.
|
|
|
Rozvrhování v distribuovaných systémech
Vyšohlíd, Jan ; Janeček, Jan (advisor) ; Yaghob, Jakub (referee)
The present work studies methods of scheduling in heterogeneous distributed systems. First there are introduced some theoretical basics which contain not only the scheduling theory itself but also the graph theory and the computational complexity theory. After that, compile-time scheduling methods and some well-known algorithms solving the problem are presented, followed by real-time scheduling basics and by classification of used methods. In the main part of the work there are proposed algorithms which respect additional restrictions. These algorithms are tested via the enclosed application and compared either to each other or to another algorithms which mostly don't respect additional restrictions. The mentioned application and the documentation for this application are a part of this work as well.
|
|
|
An application for testing automata learning algorithms
Vyšohlíd, Jan ; Krčah, Peter (referee) ; Hoffmann, Petr (advisor)
In the present work we study methods for testing regular inference algorithms. First there are introduced some theoretical basics for finite state automata and regular inference. Next we present some finite state automata learning algorithms, their principles and used testing methods, which find out algorithms quality via testing resulting automata. Following text makes the training and testing data generating process clear, describes format for saving this data and for saving finite state automata and finally the algorithms testing run alone, too. The application implementing present testing methods is a part of this work as well. It saves result statistics in chosen format. In appendices we append user and programmer documentation for this application.
|
guest ::
RSS feed.