---

GradeScope is a lightweight, machine learning–powered grade predictor. Given data on student habits and background, it intelligently estimates the final score using a trained ML model, accessible through a web-based React interface.

---

First, clone the repository and launch the backend:

cd predictor-api
./mvnw spring-boot:run

Then, launch the ML model:

cd predictor-ML
python3 -m venv .venv
source .venv/bin/activate
pip install -r requirements.txt
python app.py

Now, start the frontend:

cd frontend
npm install
npm start

Make sure to set environment variables for the backend (application.properties) and frontend (.env) where necessary. In deployment, configure these through your hosting provider.

---

Use Case

You enter a student's personal and academic information — such as gender, support received, family education level, and absences. The app sends this to the ML API, which responds with a grade prediction. Designed for educators, tutors, and students seeking insights.

---

Acknowledgements

Built using:

• Java & Spring Boot
• Python (Flask + scikit-learn)
• React + Vite
• Heroku
• Vercel

Special thanks to open data providers and educators who inspired the project structure.

---