- Homeworks could be completed on Google Colab or locally on your machine via installing Anaconda
- Lecture: Wednesday (13h-16h)
- Location: 5.56
Personal Q&A
My room is in 6.102 you can come to me with questions if needed. If you feel that you are stuck or need help with something (guidance, not pair programming) feel free to reach out to us. For additional requests email us.
Project suggestions
Please try to select a project based on your personal interest! Each project will have 3 slots which will be handed out on a ‘first come, first served’ basis! -> email: dudas.bence@ttk.elte.hu
Contacts
Ágnes Becsei: agnes.becsei@ttk.elte.hu | Task 1,2
Deutsch Norber: norbert.deutsch@ttk.elte.hu | Task 3,4
Bendegúz Borkovits: borbende@phys-gs.elte.hu | Task 5,6,7
Zoltán Kovács: k.ztoli17@gmail.com | 8,9,10
Personal project suggestions are welcome!
Project LIST
Lecture materials:
Will be updated during semester.
Schedule:
Theory
- Machine learning basics (2025.09.10)
- Course introduction
- Supervised learning
- Unsupervised learning
- Regression (2025.09.17)
- Linear regression
- Regularization
- Linear regression to classification
- Machine learning algorithms (2025.09.24)
- Suport Vector Machines
- Tree based models
- Clustering
- Deep Learning (2025.10.05)
- Multi layer perceptron
- Optimization
- Convolutional Neural Networks
- More on deep learning (2025.10.12)
- Layers for optimization
- RNN
- NLP basics
Practice
Every week in the time of the lecture, excluding theory days. This year the practice is optional but highly recommended.
Lab materials
The home works from previous year. Due to the change of the structure of this lecture these are materials for you to practice.
Number (#) | misc. | homework | info | solution |
---|---|---|---|---|
01 | data | HW 1 | UPDATED: 2023. 09. 20. | solution |
02 | dataset | HW 2 | UPDATED: 2023. 09. 20. | solution |
03 | dataset | HW 3 | UPDATED: 2023. 10. 04. | solution |
04 | dataset | HW 4 | UPDATED: 2023. 10. 11. | solution |
05 | dataset | HW 5 | UPDATED: 2023. 10. 11. | solution |
06 | code example | HW 6 | UPDATED: 2023. 10. 11. | solution |
07 | dataset, code example | HW 7 | UPDATED: 2023. 10. 25. | solution |
08 | dataset | HW 8 | N/A | solution |
09 | code example, dataset | HW 9 | UPDATED: 2023. 11. 06. | solution |
10 | Kaggle dataset | HW 10 | UPDATED: 2023. 11. 15. | solution |
11 | - | HW 11 | UPDATED: 2023. 11. 21. | Fine tuning |
12 | for extra points, GloVe file | HW 12 | UPDATED: 2023. 12. 06. | HW12 |
Requirements
Grading:
Based on two project, due dates TBD. Mid-semester project reports, Quarter-semester progress reports. Final scores: QS-PR_1 (4) + PR_1 (21) + QS-PR_2 (4) + PR_2 (21) = 50
Mark | Interval |
---|---|
5 | 42- |
4 | 35-42 |
3 | 28-35 |
2 | 21-28 |
1 | -21 |
Students also have to defend their coursework in an oral defense, in which we want to make sure you understand your submitied project.
Deadlines:
First project progress report deadline: 2025. 10. 4. 23:59 (estimated) - short email describing what you have achived thus far and zipped project files (without data if large) First project deadline: 2025. 11. 02. 23:59 - (end of fall break)
Second progress report deadline: 2025. 11. 15. 23:59 ** - same as previous Second project deadline: **2025. 12. 13. 23:59 - (end of semester)
Examplary reports of the first projects, these are some to strive for:
- SDSS best overall
- POLLUTION very good and informative EDA
- POLLUTION very good overall work
- SPOTIFY very good overall work
Send your homework to the teacher who is responsible for your project! Please make the title of your mail your Neptun-code.
Projects will be provided OR custom projects that fit your interest are welcome! You need to analyze, come up with ideas how and what to model on the data. Perform meaningful supervised learning task and unsupervised exploration. Write an informative PDF report!