beta-solver 0.0.1

A transparent, pure-math implementation of Multiple Linear Regression using the Normal Equation.

beta_solver

beta_solver is a personal project where I implement Machine Learning algorithms from scratch to understand the mathematics behind the "black box."

I am currently practicing core algorithms, starting with Linear Regression using the Normal Equation. As I learn and practice more algorithms, I will update this library to include them.

🎯 Project Goal

The goal of this project is not to replace libraries like scikit-learn, but to:

1. Demystify the math behind ML algorithms.
2. Implement pure Python/Numpy solutions without high-level wrappers.
3. Document my learning journey in code.

---

📐 Current Algorithm: Linear Regression

Right now, the library solves for the coefficient vector $\beta$ using Linear Algebra:

$$\beta = (X^T X)^{-1} (X^T Y)$$

Where:

• $X$: The input feature matrix.
• $Y$: The target vector.
• $\beta$: The resulting coefficients (Intercept + Slopes).

---

🚀 Installation

You can install beta_solver using pip:

pip install beta_solver

🤝 Contributing

This is a learning repository. If you see a way to optimize the math or make the code cleaner, feel free to open a Pull Request!

📄 License

Distributed under the MIT License. See LICENSE for more information.

👤 Author

Abhish Bondre GitHub: abhishbondre