Skip to content

Winter semester 2026/27 — 4IZ566 and 4IZ582

Both courses are taught in Czech (výuka probíhá česky, materiály jsou anglicky) and share one seminar series. Seminars are on Wednesdays in SB M15, three groups: 9:15, 11:00 and 12:30. The lecture is on Wednesday at 18:00 — every week for 4IZ566, every other week for 4IZ582 — and is given by prof. Ing. Tomáš Kliegr, Ph.D.; its schedule is in INSIS.

Timetable

Wed # Seminar (9:15 / 11:00 / 12:30) Assignment Due
23.09. 1 Seminar 1 — Foundations of the Python data stack Python Data Stack 04.10.
30.09. 2 Amazon workshop (invited, all groups)
07.10. 3 Seminar 2 — The supervised learning workflow Supervised Workflow 11.10.
14.10. 4 Seminar 3 — Language models from Python; vector representations of text LLM & Embeddings 18.10.
21.10. 5 Mid-term test (incl. programming tasks based on the assignments) · Seminar 4 — Retrieval-augmented generation and deployment RAG 08.11.
28.10. 6 Public holiday — no teaching
04.11. 7 Innovation week — no teaching
11.11. 8 ÚTIA — Deep-learning basics with PyTorch
18.11. 9 ÚTIA — Image classification
25.11. 10 ÚTIA — Detection (YOLO), segmentation (U-Net)
02.12. 11 ÚTIA — Transfer learning, face detection and recognition
09.12. 12 Term-project presentations and defences
16.12. 13 Term-project presentations and defences

Lectures are given by prof. Ing. Tomáš Kliegr, Ph.D.; times and rooms are in INSIS. 4IZ566 has a lecture every week. 4IZ582 has one every other week — 23.09., 07.10., 21.10., 02.12. and 16.12. by VŠE, 11.11. and 18.11. by ÚTIA (Institute of Information Theory and Automation, Czech Academy of Sciences).

A VŠE seminar is 45 min taught material over a Jupyter notebook, 30 min own work on the nbgrader assignment, 15 min minitest in INSIS; on 21 October the mid-term test takes the minitest's place. The ÚTIA seminars have their own format. The notebooks are handed out in class and on jupyter.vse.cz; assignments are fetched, validated and submitted through its Assignments tab under the course 4iz566-582_ZS2627.

Term project

Teams of two, milestones and the AI component on the term project page. Presentations and defences take the seminar slots on 9 and 16 December; the presentation is mandatory for every team member, and each student answers a verification question on the project.

4IZ566 versus 4IZ582

The teaching differs only in the lecture: every week for 4IZ566, every other week for 4IZ582. Seminars, assignments and the ÚTIA block are shared. In the term project the AI component is mandatory for 4IZ582 and an optional extension for 4IZ566.

  • 4IZ460 — Advanced approaches to knowledge discovery from databases (INSIS)
  • 4IZ569 — Cloud Data Mining (INSIS)

Consultations with Ing. Lukáš Sýkora, Ph.D. are held on Microsoft Teams, by prior e-mail arrangement — see Teachers.