📊 Linear Regression – Car Price Prediction

🌟 Highlights

• Machine Learning from Scratch – No Scikit-Learn or built-in regression functions.
• Gradient Descent Optimization – Supports multiple learning rates to optimize training.
• Data Normalization & Feature Scaling – Ensures stable convergence during training.
• Model Persistence – Saves trained parameters for future predictions.
• Visualization – Plots regression results and dataset for better interpretation.
• Command-Line Interface – Train and predict using simple CLI commands.

---

ℹ️ Overview

This project implements a simple linear regression model to predict car prices based on mileage. Using a gradient descent algorithm, the model optimizes theta0 (intercept) and theta1 (coefficient) to fit a linear function of the form:

estimatePrice(mileage) = theta0 + (theta1 * mileage)

Key Features: ✅ Implements gradient descent to iteratively improve accuracy. ✅ Normalizes features for better convergence. ✅ Saves trained model parameters in model.json for future predictions. ✅ Provides visualization tools to analyze dataset distribution and model fit. ✅ Supports command-line input for training and predictions.

---

🚀 Installation & Setup

1. Clone the Repository

git clone https://github.com/LuckyIntegral/linear_regression.git
cd linear_regression

2. Install Dependencies

pip install -r requirements.txt

---

🔄 Training the Model

Run the training script to optimize model parameters:

python linear_regression/train.py

This will generate a model file (model.json) storing the trained values of theta0 and theta1.

---

🔮 Making Predictions

Use the trained model to predict the price of a car based on mileage:

python linear_regression/predict.py

This script prompts the user for mileage input and returns an estimated car price.

---

📊 Data Visualization

Before training, analyze the dataset using built-in visualization tools:

Generate a Regression Plot

python linear_regression/predict.py

This will display a graph of the dataset, the regression line, and the user’s prediction.

---

🧪 Project Structure

.
├── README.md
├── data
│   └── data.csv
├── linear_regression
│   ├── predict.py
│   └── train.py
└── requirements.txt

---

✨ Future Improvements

• Add Polynomial Regression for non-linear relationships.
• Support Multiple Linear Regression with additional features.
• Improve Precision Score Calculation.

---

🎓 Author

Vitalii Frants 📍 42 Vienna – AI & Algorithms 👉 GitHub

---

💎 Ready to Predict Car Prices? Try it now!