mini_ml

A minimalistic, educational machine learning library implemented from scratch in Python. This project is designed for learning and experimentation, providing clear, well-documented code for core ML algorithms and utilities.

Features

• Core ML algorithms: KNN, Linear Regression, Logistic Regression, and more
• Utility modules: Activation functions, optimization, preprocessing, statistics, and linear algebra
• Simple, readable code for educational purposes
• Comprehensive tests for each module

Installation

Linux (Recommended)

git clone https://github.com/YOUR_USERNAME/mini_ml.git
cd mini_ml
python3 -m venv venv
source venv/bin/activate
pip install -r requirements.txt

Windows

• Use Git Bash or WSL for best compatibility.
• Follow the Linux instructions above inside your terminal.

Usage

Each module in src/ has its own documentation and usage guide:

• Activation Functions
• KNN
• Linear Algebra
• Linear Model
• Logistic Model
• Optimization
• Preprocessing
• Statistics

Each guide covers:

• Theory, math, and intuition
• Main functions/classes to use
• Short code examples

Example: Using KNN

from src.knn import KNNClassifier

X_train = [[0, 0], [1, 1], [2, 2]]
y_train = [0, 1, 1]

knn = KNNClassifier(k=3)
knn.fit(X_train, y_train)
pred = knn.predict([[1.5, 1.5]])
print(pred)

Contributing

1. Fork the repository and create a new branch from main.
2. Add or improve features. Write tests for your code in the tests/ folder, following the existing style.
3. Run all tests with:

   uv sync
   uv run -m pytest

   Ensure all tests pass. Add a screenshot of your terminal showing all tests passing to your PR.
4. Use semantic commit messages.
5. Open a pull request.

See CONTRIBUTING.md for more details.

---

This project is for educational purposes. Contributions and feedback are welcome!