HaywhyCoder/Lagos-Housing-Predictor

A machine learning practice project designed for the NSBE UNILAG Machine Learning Skills Lab Training Program that predicts housing prices in Lagos, Nigeria. # 🏡 Nigerian Housing Predictor A machine learning practice project designed for the **NSBE UNILAG Machine Learning Skills Lab Training Program** to teach beginner machine learning concepts. ## 🌐 Live Demo **🚀 Try the app now:** Lagos Housing Predictor - Live Demo ## 📋 Project Overview This project demonstrates end-to-end machine learning workflow by predicting housing prices in Lagos, Nigeria. It serves as a hands-on learning tool for trainees to understand data science concepts... ## 🎯 Learning Objectives This practice project helps trainees learn: - Data preprocessing and feature engineering - Machine learning model training and evaluation - Model serialization and deployment - Building interactive web applications with Streamlit - Python programming best practices - Project structure and organization ## 🚀 Features - **Interactive Web Interface**: User-friendly Streamlit application for price predictions - **Real-time Predictions**: Get instant housing price estimates based on input parameters - **Multiple Input Parameters**: - Number of bedrooms, bathrooms, and toilets - Parking spaces - House type (Detached Duplex, Terraced Bungalow, etc.) - Location within Lagos - **Pre-trained Model**: Uses a saved machine learning pipeline for predictions ## 📁 Project Structure ``` Lagos-Housing-Predictor/ ├── data/ │ └── nigeria_houses_data.csv # Training dataset ├── models/ │ └── my_prediction_pipeline.pkl # Pre-trained ML model ├── notebooks/ │ └── Nigerian_Housing.ipynb # Jupyter notebook for analysis ├── src/ │ ├── app.py # Streamlit web application │ └── main.py # Launcher script ├── requirements.txt # Python dependencies └── README.md # Project documentation ``` ## 🛠️ Installation & Setup ### Prerequisites - Python 3.7 or higher - pip package manager ### Steps 1. **Clone/Download the project** ```bash # Clone from GitHub git clone github.com …