--- title: StudentMarksPrediction emoji: 🚀 colorFrom: red colorTo: red sdk: docker app_port: 8501 tags: - streamlit pinned: false short_description: Predicting student marks with a Gradient Boosting model. license: mit --- # 🎓 Student Grade Prediction App This project is a simple and clean machine learning application that predicts a student's marks based on two inputs: - **Number of courses** - **Study time per day (hours)** The app uses a **Gradient Boosting Regression model** trained on a small dataset. It is built with **Streamlit** and deployed on **HuggingFace Spaces**. --- ## 🚀 Features - Easy-to-use Streamlit interface - Real-time grade prediction - Machine learning model saved as `.pkl` for fast loading - Very small and lightweight app - Works directly in the browser --- ## 🧠 Model Information - **Model type:** GradientBoostingRegressor - **Library:** scikit-learn - **Target variable:** Marks - **Most important feature:** Study time per day The model achieves strong performance with a low RMSE and a very high R² score. Note: The dataset is very small (100 samples), so the model learns only simple patterns. Values that are outside the range of the training data (for example, 12 hours of study or more courses than in the dataset) can produce unrealistic predictions. --- ## 📂 Project Structure ├── app.py # Streamlit web app ├── gb_model.pkl # Trained machine learning model ├── requirements.txt # Dependencies for HuggingFace Space └── README.md # This file yaml Code kopieren --- ## 🛠️ How to Run the App Locally 1. Install requirements: pip install -r requirements.txt markdown Code kopieren 2. Run Streamlit: streamlit run app.py yaml Code kopieren The app will open in your browser. --- ## 📁 Dataset Source The dataset contains: - number of courses - daily study time - final marks It is publicly available online.