This repository provides a comprehensive collection of seminar materials and assignments for a machine learning course, covering foundational to advanced topics. It is designed for students and practitioners seeking to deepen their understanding and practical skills in machine learning through structured learning and hands-on exercises.

How It Works

The repository organizes learning materials by seminar topic, with each seminar typically including theoretical explanations and practical code examples, often using Python libraries like scikit-learn and PyTorch. Assignments are provided with clear requirements, deadlines, and evaluation criteria, encouraging practical application of learned concepts. The structure facilitates a step-by-step progression through various ML paradigms, from linear models and kernel methods to deep learning architectures and specialized topics like NLP and time series.

Quick Start & Requirements

• Install/Run: Primarily involves cloning the repository and running provided Jupyter notebooks (.ipynb files).
• Prerequisites: Python 3.x, Jupyter Notebook, and common ML libraries (scikit-learn, PyTorch, pandas, numpy, matplotlib). Specific seminars might require additional libraries like torch, tensorboard, xgboost, catboost, pyserini.
• Resources: Requires a standard development environment with Python and necessary libraries installed. Some assignments may benefit from GPU acceleration for deep learning tasks.
• Links:
  • 2020-2021 Playlist: https://www.youtube.com/playlist?list=PLk4h7dmY2eYHHTyfLyrl7HmP-H3mMAW08
  • 2021-2022 Playlist: https://youtube.com/playlist?list=PLk4h7dmY2eYFmowaPqjFDzSokiiLq5TkT
  • K.V. Vorontsov's Lecture Course: http://www.machinelearning.ru/wiki/index.php?title=Машинное_обучение_%28курс_лекций%2C_К.В.Воронцов%29

Highlighted Details

• Covers a wide range of ML topics, from linear models and kernel methods to deep learning (CNNs, RNNs), NLP (Word2Vec, Transformers), and reinforcement learning.
• Includes practical assignments with clear evaluation criteria, encouraging hands-on learning and skill development.
• Provides links to relevant YouTube playlists and external course materials for supplementary learning.
• Features code examples and theoretical explanations for each seminar topic, facilitating a structured learning path.

Maintenance & Community

• The repository is maintained by Andriygav.
• Contact email for inquiries: grabovoy.av@phystech.edu.

Licensing & Compatibility

• The repository itself does not explicitly state a license. Code within notebooks is generally compatible with standard Python environments.

Limitations & Caveats

• The content is primarily in Russian.
• Some assignments have specific submission deadlines that have likely passed.
• The repository focuses on educational materials rather than a deployable ML library.