🚀 A Machine Learning-based web application that predicts temperature using real-time weather parameters. Built using Flask, Random Forest, and deployed on Render.
👉 https://weather-temperature-prediction-ml-project.onrender.com
📸 Preview
After Prediction
This project predicts the temperature (°C) based on multiple weather features such as:
- Wind Speed
- Wind Direction
- Pressure
- Humidity
- Cloud Coverage
- Visibility
- Sunrise Time
The model is trained using a Random Forest Regressor and deployed as a web application for real-time predictions.
-
Algorithm: Random Forest Regressor
-
Dataset Size: ~130,000 rows
-
Preprocessing:
- Handling missing values
- Feature engineering (sunrise → minutes)
- One-hot encoding (wind direction)
- Accuracy: 71.50%
- R² Score: 0.715
- RMSE: 5.16
| Category | Tools |
|---|---|
| Backend | Flask |
| ML Model | Scikit-learn |
| Data Processing | Pandas, NumPy |
| Deployment | Render |
| Server | Gunicorn |
| Frontend | HTML, CSS, Bootstrap |
✅ Real-time temperature prediction ✅ Clean and responsive UI ✅ Smart input suggestions ✅ Auto-fill feature for quick testing ✅ Model performance metrics displayed ✅ Fully deployed web app
project/
│
├── app.py
├── model.pkl
├── columns.pkl
├── accuracy.pkl
├── r2.pkl
├── rmse.pkl
├── requirements.txt
├── Procfile
│
└── templates/
└── index.html
- User inputs weather parameters
- Data is preprocessed (encoding + transformation)
- Model predicts temperature
- Results + metrics displayed on UI
git clone https://github.com/YOUR_USERNAME/weather-ml.git
cd weather-ml
pip install -r requirements.txt
python app.pyThis project is deployed on Render using:
- Gunicorn (WSGI server)
- Flask backend
- Pre-trained ML model
- 📊 Add visualization graphs
- 🌍 Integrate live weather API
- 🤖 Add chatbot interface
- 📱 Improve UI/UX with animations
- 🔐 Add authentication system
Gowtham P B.Tech CSE (AI) Aspiring ML Engineer 🚀
Give it a ⭐ on GitHub and share it!

