Linear Regression Using Python

A clean, minimal implementation of Linear Regression from scratch using NumPy. This project is built for learning and quick experimentation, with ready-to-run notebooks and sample data.

---

✨ Highlights

• Pure NumPy implementation (no scikit-learn)
• Simple API with LinearModel class
• Example notebooks included
• Works with your own CSV/Excel datasets

---

🚀 Quick Start

1. Clone the repo

2. Install dependencies

pip install numpy pandas matplotlib xlrd

3. Run a notebook

Open and run:

• example_implementation.ipynb
• main.ipynb

---

Basic Usage

from linear_model import LinearModel
import numpy as np

# X shape: (m, n) and y shape: (m, 1)
X = np.array([[1.0], [2.0], [3.0], [4.0]])
y = np.array([[1.2], [1.9], [3.1], [3.9]])

model = LinearModel(num_features=1)
losses = model.train(X, y, iterations=2000, lr=0.01)

y_pred = model.forward_pass(X)
print(y_pred)

---

📊 Using the Sample Data

Sample files are included:

• chirps.csv
• chirps.xls

You can load the Excel file like this:

import pandas as pd

df = pd.read_excel("chirps.xls", engine="xlrd")

---

📁 Project Structure

.
├── linear_model.py                 # Core Linear Regression implementation
├── example_implementation.ipynb    # Step-by-step walkthrough
├── main.ipynb                       # Main usage notebook
├── chirps.csv                       # Sample dataset (CSV)
├── chirps.xls                       # Sample dataset (Excel)
└── README.md                        # Project guide

---

🧠 How the Model Works

The model minimizes Mean Squared Error using gradient descent:

$$ J(\theta) = \frac{1}{2m} \sum_{i=1}^{m} (\hat{y}^{(i)} - y^{(i)})^2 $$

Parameters are updated iteratively using:

$$ \theta := \theta - \alpha \nabla J(\theta) $$

---

🔧 Tips

• Scale your features if values are large.
• Increase iterations for better convergence.
• Tune lr (learning rate) if loss is unstable.

---

Author

Created by Rehnoor Aulakh.

If you find this helpful, feel free to star the repo and share it.