IamMishael06/marsHouse

ML project that gives an annual average rent prices of houses in Nigeria. Incorporating data science and ML models ## MarHouse – House Rent Prediction in Naira 📌 Overview MarHouse is a machine learning project that predicts the annual rent of houses in Nigeria (in Naira) based on various features like location, property type, and amenities. The project includes: * Data collection from a real estate site using a custom scraper (via ScraperAPI). * Data cleaning & feature engineering to turn messy, unstructured housing listings into meaningful data. * Model training using `RandomForestRegressor` for accurate predictions. * Deployment-ready backend with FastAPI. * Containerize with Docker and deploy. ### ⚙️ Tech Stack * Python – Core language * Pandas / NumPy – Data manipulation * Scikit-learn – Machine learning model and preprocessing * ScraperAPI – Data scraping * Joblib – Model saving/loading * Django – Web framework for deployment ### 📊Features & Input Variables | Feature | Description | | ---------------------- | ------------------------------------------ | | Bedrooms | Number of bedrooms | | Bathrooms | Number of bathrooms | | Toilets | Number of toilets | | Parking lot | Available parking spaces | | Extras | Additional features count | | Serviced | Whether the house is serviced (1/0) | | Stable Electricity | Whether electricity supply is stable (1/0) | | State | State where the property is located | | Town | Town where the property is located | | Title (Apartment Type) | e.g., Duplex, Bungalow, Flat | ### 🧠 Model Details Algorithm: RandomForestRegressor Encoding: * OneHotEncoder – For state and town * TargetEncoder – For title (apartment type) * Pipeline: Combined preprocessing and model in a single scikit-learn pipeline. * Output: Predicted annual …