A modern web application that predicts real estate prices in Nepal's Kathmandu Valley using machine learning. Built with a Next.js frontend and a Flask backend, this app provides real-time price estimates based on property details.
🔗 Backend GitHub Repository
- Backend: NepalEstate API
- Price Prediction: Predict property prices based on:
- Number of floors
- Area (Anas)
- Road width (ft)
- City (Kathmandu, Bhaktapur, or Lalitpur)
- Road type (Blacktopped, Gravelled, or Soil Stabilized)
- Responsive Design: Built with Next.js and Tailwind CSS for modern, mobile-friendly UI
- Interactive Form: Simple and intuitive user inputs
- Real-time Predictions: Instantly receive price estimates using a trained
MLPRegressormachine learning model after evaluating different models
- Frontend: Next.js
- Backend: Flask, scikit-learn
- Machine Learning: MLPRegressor (
scikit-learn)
🔗 GitHub Repository
- Backend: NepalEstate API
To run the project locally on your device:
- Node.js (LTS version recommended)
- Python 3.13 or compatible
- Git
# Clone the repository
https://github.com/080bct12alex/NepalEState.git
# Navigate to Flask server
cd api
# Create and activate virtual environment
python -m venv venv
# Windows
venv\Scripts\activate
# macOS/Linux
source venv/bin/activate
# Install dependencies
pip install -r requirements.txt
# Run Flask server
python app.py
The Flask backend runs at: http://127.0.0.1:5001
💻 Frontend Setup
# Install dependencies
npm install
# Start development server
npx next dev
The Next.js frontend runs at: http://localhost:3000
🚀 Usage
Make sure both frontend and backend servers are running.
Open your browser and go to http://localhost:3000.
Fill in the property details:
Number of floors
Area in Anas
Road width in ft
Select a city
Select road type
Click the "Predict Price" button.
View the estimated property price on the screen.