ML Practices

A collection of educational machine learning notebooks, homework assignments, and small experiments.

This repository is my personal workspace for practicing data analysis, classical machine learning, neural networks, embeddings, semantic search, and AI agents.

Repository structure

ML-Practices/
├── yandex/
│   ├── agents_week/
│   │   └── submission.ipynb
│   ├── vsosh/
│   │   ├── vsosh_ht_1.ipynb
│   │   ├── vsosh_ht_2.ipynb
│   │   ├── vsosh_ht_3.ipynb
│   │   ├── vsosh_ht_4.ipynb
│   │   └── vsosh_ht_5.ipynb
│   └── ya_spec/
│       └── ya_spec_1.ipynb
└── README.md

Contents

yandex/agents_week

Notebook with a homework assignment on AI agents. It includes practice with LLM calls, tools, tool tracing, and a small product-catalog environment for agent experiments.

yandex/vsosh

A set of machine learning homework notebooks:

Notebook	Topic
vsosh_ht_1.ipynb	Exploratory data analysis and visualization on the Netflix dataset
vsosh_ht_2.ipynb	Heart Disease classification with EDA and model comparison
vsosh_ht_3.ipynb	Regression model comparison on the Diamonds dataset
vsosh_ht_4.ipynb	Text classification and analysis of bank customer reviews
vsosh_ht_5.ipynb	Word embeddings, FastText training, and semantic search for similar questions

yandex/ya_spec

Notebook on introductory neural networks. It includes work with MNIST, PyTorch, training loops, validation, testing, and classification metrics.

Main topics

• Exploratory data analysis
• Data visualization
• Classical machine learning
• Classification and regression
• Text preprocessing and vectorization
• Word embeddings and semantic search
• Neural networks with PyTorch
• AI agents and tool usage
• Practical work with Jupyter notebooks

Technologies

The repository uses different parts of the Python ecosystem, including:

• Python
• Jupyter Notebook
• NumPy
• pandas
• matplotlib
• scikit-learn
• PyTorch
• torchvision
• gensim
• FAISS
• LangChain

How to run

Clone the repository:

git clone https://github.com/amemeansrain/ML-Practices.git
cd ML-Practices

Create and activate a virtual environment:

python -m venv .venv

# Windows
.venv\Scripts\activate

# macOS / Linux
source .venv/bin/activate

Install the packages needed for the notebook you want to run.

For general notebook-based ML tasks:

pip install numpy pandas matplotlib scikit-learn jupyter

For neural-network notebooks:

pip install torch torchvision

For embeddings and semantic search:

pip install gensim faiss-cpu

For AI agents:

pip install langchain-openai langchain-core

Start Jupyter:

jupyter notebook

Then open the needed notebook from the yandex/ directory.

Notes

This repository is mainly educational. The notebooks are used for practice, experimentation, and tracking progress while studying machine learning, data analysis, neural networks, AI systems, and applied Python tasks.