Practical Machine Learning in Python
General data
Course ID: | 2400-ZEWW758 |
Erasmus code / ISCED: |
14.3
|
Course title: | Practical Machine Learning in Python |
Name in Polish: | Praktyczny Machine Learning w Pythonie |
Organizational unit: | Faculty of Economic Sciences |
Course groups: |
(in Polish) Przedmioty kierunkowe do wyboru - studia II stopnia EP - grupa 4 (1*30h) (in Polish) Przedmioty kierunkowe do wyboru - studia II stopnia IE - grupa 2 (2*30h) (in Polish) Przedmioty kierunkowe do wyboru- studia I stopnia EP (in Polish) Przedmioty wyboru kierunkowego dla studiów licencjackich EM (in Polish) Przedmioty wyboru kierunkowego dla studiów licencjackich IE (in Polish) Przedmioty wyboru kierunkowego dla studiów licencjackich MSEM |
ECTS credit allocation (and other scores): |
3.00
|
Language: | Polish |
Type of course: | optional courses |
Short description: |
The aim of the course is to present the latest machine learning methods, using Python programming language. In order to fully illustrate the machine learning issues, all three types of analysis will be presented: unsupervised, supervised and reinforcement learning. However, the main focus will be on predictive models, both classification and regression. Classes will focus on the formation of an intuitive understanding of the algorithms discussed, their strengths and weaknesses and the acquisition of practical skills to use them. |
Full description: |
•Introduction to machine learning ◦Data science, Data Mining, Deep Learning, Big Data and Machine learning. ◦Machine learning for business ◦Machine learning as a function •Supervised learning ◦Regression and classification. Objective function. ◦Bias variance dillema. ◦Linear regression and logistic regression ◦Decision tress. ◦Simple decision trees. ◦Random Forest. ◦Boosting. ◦Neural networks. •Unsupervised learning ◦Clustering. ◦Not only K-Means: comparison of different clustering methods. ◦Curse of multidimensionality ◦Dimensions reduction. ▪Principal component analysis (PCA) ▪Self-organizing map (SOM) ▪t-distributed stochastic neighbour embedding (t-SNE) •Reinforcement Learning |
Bibliography: |
Harrington, Peter. Machine learning in action. Vol. 5. Greenwich, CT: Manning, 2012 |
Learning outcomes: |
Knowledge Student knows methods of predictive modelling. Student knows and understands methods based on decision trees and neural networks. Student knows the sources of obtaining large data sets. Student knows methods of using Python for data analysis. Student knows applications of presented statistical methods and can create market analyses at work or for the needs of his own company. Skills Student can choose appropriate modelling method for a given problem. On the basis of acquired knowledge, student can formulate his/her own opinion and apply the theoretical knowledge to description and analysis of economic phenomena. Student can look for and find data sets, apply predictive modelling and prepare description of performed analysis. Social skills The practice of using Python programming language allows to increase the skills of independent learning and increases the competences in object-oriented programming. The exercises and modelling practices carried out during the course allow students to be critical of the results obtained in scientific work. |
Assessment methods and assessment criteria: |
Final project (60%) Exam (test) (40%) An extra possibility to pass the course is to participate and achieve a good result in the data analysis competition (e.g. kaggle). Details will be given at the first meeting. |
Classes in period "Summer semester 2024/25" (past)
Time span: | 2025-02-17 - 2025-06-08 |
Go to timetable
MO TU KON
W TH FR |
Type of class: |
Seminar, 30 hours
|
|
Coordinators: | Maciej Wilamowski | |
Group instructors: | Maciej Wilamowski | |
Students list: | (inaccessible to you) | |
Credit: |
Course -
Grading
Seminar - Grading |
Copyright by University of Warsaw.