gexinyang0-boop/car-price-prediction

Predicting used car prices in Morocco using machine learning and a Streamlit dashboard # Car Price Prediction A data science project that predicts used car prices in Morocco. It includes data cleaning, exploratory data analysis (EDA), machine learning model training and comparison, and an interactive Streamlit dashboard. ## Project Overview The goal is to predict the price of a used car from its features (such as year, mileage, fuel type, gearbox, and fiscal power). Ten regression models were trained and compared. A Streamlit web app lets users explore the data and make live price predictions. ## Results Ten models were compared using R² (higher is better). The best model was **Random Forest**. | Model | R² | |---|---| | Random Forest | 0.62 | | Extra Trees | 0.61 | | HistGradientBoosting | 0.56 | | KNN Regression | 0.50 | | Gradient Boosting | 0.47 | | Ridge Regression | 0.42 | | Linear Regression | 0.39 | ## Project Structure ``` car-price-prediction/ ├── Project（Advanced_data_science）.ipynb # Main notebook: cleaning, EDA, modelling ├── Advanced Data Science Report.pdf # Full written report ├── data/data.zip # Datasets (raw + cleaned), zipped ├── app/ # Streamlit dashboard │ ├── app.py # Home page │ └── pages/ # Overview, EDA, Model Comparison, Predict, Map ├── models/ # Saved model files ├── outputs/ # Charts and result files ├── requirements.txt # Python packages needed └── README.md ``` ## How to Run 1. Clone this repository: ``` git clone github.com cd car-price-prediction ``` 2. Unzip the datasets: ``` unzip data/data.zip -d . ``` This gives you `cars_dataframe.csv` (raw) and `cars_cleaned.csv` (cleaned). 3. Install the required packages: ``` pip install -r requirements.txt ``` 4. Run the Streamlit dashboard: ``` streamlit run app/app.py ``` ## Notes - The best …