Data Analysis and Visualization in Python
IZV Acad. year 2023/2024 Winter semester 4 credits
The aim of the course is to acquaint students with the problems of data acquisition, processing, analysis and visualization using the cross-platform scripting language Python. It has a sophisticated ecosystem offering a rich spectrum of extension libraries, either in the form of native code or in terms of performance of efficient extensions implemented in C / C ++.
During the lectures students will learn Python constructs, methods of data acquisition, storage and manipulation, possibilities of advanced computations in numerical and symbolic level and visualization of acquired data. In this course, students will also gain an overview of the properties of techniques for advanced analysis of data dependencies and their applications for various data. Finally, Python will be expanded to include custom designs and techniques to effectively overcome the disadvantages of the interpreted language for performance-oriented applications. In the practical part (project), students will go through all stages of large data processing - from the design stage, through processing to subsequent analysis and visualization.
Language of instruction
- 26 hrs lectures
- 13 hrs projects
- 100 pts projects
The aim of the course is to acquaint students with the issue of data acquisition, processing and analysis in Python. Within the course, Python will be introduced as a tool for efficient data manipulation.
Students will gain a general overview of basic and advanced methods of data analysis and basic and advanced aspects of Python, which they will learn to use with modern mathematical libraries and libraries for advanced data analysis and modeling. They will understand how the techniques implemented in these libraries work in general and learn what technique is appropriate for what data.
In addition to general knowledge of basic data processing techniques, the student will gain an overview of the effective execution of critical parts of the program, extension of the language with its own modules written in C / C ++ or the problematic of installing libraries in an isolated environment or containers.
At the end of the course students should understand how to effectively obtain, analyze and visualize data of various extent. The knowledge can then be used to solve non-trivial engineering and scientific tasks or to evaluate data for management and decision-making purposes.
Why is the course taught
The aim of the course is to increase students' employability in the labor market, where data operations at various levels are becoming increasingly important, especially considering the connection with artificial intelligence. The aim is to show how to effectively use existing frameworks and techniques across the entire spectrum of data processing and analysis processes, from obtaining the data itself to automatically generating a report presenting the results of data analysis, and to be able to apply them in practice in the commercial sphere and while working on bachelor and master theses.
Prerequisite knowledge and skills
Basic knowledge of imperative programming and algorithmization, knowledge of basic concepts and operations from linear algebra (vectors, work with matrices, linear operations, etc.) and statistics.
- IZP - Introduction to Programming Systems
- ILG - Linear algebra
- IPP - Principles of Programming Languages
- IPT - Probability and Statistics
- Mark Pilgrim: Ponořme se do Pythonu 3 (ISBN: 978-80-904248-2-1, dostupné online)
- Jake VanderPlas: Python Data Science Handbook (ISBN: 978-1-491-91205-8, dostupné v online zdrojích knihovny)
- Samir Madhavan: Mastering Python for Data Science (ISBN: 978-17-843901-5-0)
- Robert Johansson: Numerical Python (2019, ISBN: 978-1-4842-4245-2)
Syllabus of lectures
- Introduction to Language I
- Introduction to Language II
- Data acquisition and data persistence
- Effective implementation of operations over n-dimensional fields
- Tools for advanced data manipulation
- Basic approaches to data visualization
- Basic methods of data and data dependency analysis
- Advanced approaches to data visualization
- Advanced methods of data and data dependency analysis
- Work with image data and possibilities of data presentation
- Advanced operations over time series
- Symbolic domain calculations
- Code acceleration capabilities for HPC needs
Syllabus - others, projects and individual work of students
The aim of the project is to create a script that obtains data from publicly available sources, analyzes and presentes it in the form of a report. Project evaluation will take into account the quality of the code, the resulting analysis and the generated reports.
The evaluation is based on individual implementation of a project whose implementation consists of three parts (data acquisition, data preprocessing and analysis, report generation). Each part will be evaluated separately. Students will receive feedback on their work, which they will incorporate into the final solution. The first two parts are submitted during the semester and can be awarded up to 20 points each. Up to 60 points can be earned for the final solution.
Minimum of 50 points earned. At least 2 points from each part of the project.
|1., 2., 3., 4., 7., 10., 13. of lectures
|E104 E105 E112
|5., 6., 8., 9., 11., 12. of lectures
|E104 E105 E112
Course inclusion in study plans