Installation Guide

Make sure the following are installed and accessible from your terminal before proceeding:

• Python 3.10 or higher
• Flutter SDK (latest stable channel)
• PostgreSQL 12+ (for production) or SQLite (default for development)
• An IDE like VS Code or Android Studio
• A connected Android device or emulator

Open your terminal and run the commands below one after another.

1. Move into the backend folder: cd backend
2. Install all Python dependencies: pip install -r requirements.txt
3. Apply database migrations: python manage.py migrate
4. Create an admin user (optional): python manage.py createsuperuser
5. Start the development server: python manage.py runserver 0.0.0.0:8000

Once running, the API will be available at http://localhost:8000.

The repository already ships with pre-trained models inside backend/assessment/models/, but if you wish to retrain them on the included dataset:

1. Stay inside the backend folder.
2. Run the training script: python assessment/train_model.py

This will load the PCOS dataset from backend/dataset/PCOS_data.csv, train both Random Forest and Logistic Regression models, pick the best performer automatically, and save it back to the models folder. Expected accuracy is around 83% on cross-validation.

Open a new terminal window for the mobile app.

1. Move into the Flutter project folder: cd pcos_tracker
2. Install Flutter packages: flutter pub get
3. Open lib/config/constants.dart and update AppConstants.baseUrl. If you are running on a physical phone, use your laptop's local IP (for example http://192.168.1.5:8000) instead of localhost.
4. Connect a device or start an emulator.
5. Run the app: flutter run

1. Register a new account from the signup screen.
2. Log in to receive your JWT tokens automatically.
3. Log your menstrual cycle, daily symptoms, mood, exercise, and nutrition.
4. Open the Assessment screen and enter your health parameters to receive an AI-generated PCOS risk score along with personalized recommendations.
5. Browse the Insights tab to view your cycle regularity, mood distribution, and symptom trends.

• App can't reach backend: Ensure your laptop and phone are on the same Wi-Fi network, and the base URL points to your laptop's IP address.
• Migration errors: Delete db.sqlite3 and re-run python manage.py migrate.
• ML model file missing: Run python assessment/train_model.py once to regenerate the saved model.
• Flutter package issues: Run flutter clean followed by flutter pub get.

If you face any issues during installation or want a one-on-one walkthrough of the codebase before your viva, our project setup and mentorship service covers everything from running the project to defending it in front of your panel. Reach out through the CodeAj Marketplace to get full support until submission.