42 ft_linear_regression 🚗📈

Overview ✨

This project is an machine learning toolset for performing linear regression analysis on car mileage and price data. Linear regression is a foundational technique in artificial intelligence and machine learning, used for predicting continuous values. This project covers the full pipeline: data scraping, cleaning, model training, prediction, and evaluation. The codebase is modular, allowing users to generate datasets, train a model, make predictions, and evaluate model precision.

---

Features 🚀

• 🤖 ML Powered: Utilizes linear regression, a core machine learning algorithm, to model and predict car prices.
• 🕸️ Dataset Generation: Scrape car mileage and price data from the web, clean and save as CSV.
• 🤖 Model Training: Train a linear regression model to predict car prices based on mileage.
• 🔮 Prediction: Predict car prices for given mileage using trained model parameters.
• 📊 Precision Calculation: Evaluate model performance with MSE, RMSE, and R-squared metrics.
• 📉 Data Visualization: Plot data points and regression line.

---

Project Structure 🗂️

42_ft_linear_regression/
├── data/                # Datasets and model parameters (CSV files)
├── requirements.txt     # Python dependencies
└── src/
    ├── dataset_generator.py      # Scrape and clean data
    ├── precision_calculator.py   # Evaluate model precision
    ├── predictor.py              # Predict price from mileage
    ├── trainer.py                # Train linear regression model
    └── utils/
        ├── data_plotter.py       # Plotting utilities
        ├── errors.py             # Error handling
        ├── file_utils.py         # Data/model I/O
        ├── linear_regression.py  # Core regression logic
        ├── scraper.py            # Web scraping logic
        ├── url.py                # URL handling
        ├── validators.py         # CSV/data validation
        └── __init__.py

---

Installation ⚙️

1. Clone the repository:

   git clone <repo-url>
   cd 42_ft_linear_regression

2. Install dependencies:

   pip install -r requirements.txt

---

Usage 🛠️

1. Generate Dataset 🕸️

Scrape and clean car data, then save as CSV:

python src/dataset_generator.py -p data/data_original.csv -n 1000

2. Train Model 🤖

Train a linear regression model on your dataset:

python src/trainer.py -d data/data_original.csv --save_thetas -t data/thetas.csv --show_data

3. Predict Price 🔮

Predict the price for a given mileage:

python src/predictor.py -p data/thetas.csv

4. Evaluate Model Precision 📊

Calculate MSE, RMSE, and R-squared for your model:

python src/precision_calculator.py -d data/data_original.csv -t data/thetas.csv

---

Dependencies 📦

• pandas
• numpy
• matplotlib
• tqdm
• requests
• beautifulsoup4
• lxml

Install all dependencies with:

pip install -r requirements.txt

---

How It Works 🔄

1. Data Acquisition: 🕸️ Scrape car data → Clean and optimize dataset → Save as CSV.
2. Model Training: 🤖 Load dataset → Train linear regression model → Save thetas.
3. Prediction: 🔮 Load thetas → Input mileage → Predict price.
4. Evaluation: 📊 Load dataset and thetas → Calculate precision metrics.

---