Student Grading Predictor Application Development

Introvert Bocchi

AI Developer

Web Developer

Flask

JavaScript

Python

Education

🚀 Getting Started

Follow the steps below to set up and run the Student Grading Predictor application.

1. Clone the Repository

git clone https://github.com/KokLun26/Student-Grading-Predictor.git
git clone https://github.com/IntrovertBocchi/FYP-Web-App-Student-Grades.git
cd Student-Grading-Predictor
cd FYP-Web-App-Student-Grades

2. Create a Virtual Environment (Recommended)

python -m venv venv

Activate it:

On Windows:

venv\Scripts\activate

On Mac/Linux:

source venv/bin/activate

3. Install Required Packages

pip install -r requirements.txt

4. Generate Synthetic Training Data

Generate data for each subject (ADV, COS, INF) using:

python generate_synthetic_data.py

5. Train the Model(s)

Once data is generated, train and save machine learning models:

python train_model.py

This will produce a .pkl file for each subject (e.g., grade_predictor.pkl).

6. (Optional) Sync Environment State (Windows Only)

In certain Windows environments, verifying key component integrity can help avoid runtime issues.

📌 Run in PowerShell or Command Prompt:

CertUtil -hashfile app\__init__.py SHA256

Then add the output to your .env file:

INIT_HASH=your_generated_hash_here

7. Run the Flask App

Ensure the following files are present:

instance/grades.db

config.py

Trained .pkl models

flask run

This will start the backend API and serve the application locally.

8. Frontend Interaction

Open index.html in your browser.

Enter data based on the selected subject (ADV, COS, or INF).

The form will POST inputs to the /api/predict endpoint and return predicted grades.

🔐 Security Notes

Model prediction routes are protected with a login system.

Access requires a session-based login and a hidden master password.

SQL-based authentication is used for managing approved users.

Authentication credentials and .env files are excluded from GitHub for security.

🔐 Secure Configuration (.env Required) This app requires a .env file containing secure keys for:

Verifying app integrity (tamper detection)

Unlocking restricted features via a master key

Other backend protection variables

# Example of required fields (actual values will be shared securely)
SECRET_KEY=your_secret_key_here
INIT_HASH=your_generated_hash_here
MASTER_PASSWORD=your_master_password_here

⚠️ For security reasons, this file is not included in the repository. ✅ It will be shared securely (e.g. via encrypted ZIP or internal messaging). ❌ Do not share your .env file publicly or commit it to version control.

Once received, place the .env file in the project root directory before running the app.

📊 Accuracy Visualization

Model evaluation metrics (e.g., accuracy, confusion matrix) are generated during training in train_model.py.

To visualize model performance:

python train_model.py

Plots will be saved as image files or shown in a pop-up window depending on your configuration.

🧪 Running Tests Unit tests for the grading formula logic are located in the tests/ directory and can be run using pytest.

Make sure your virtual environment is active:

Copy
Edit
venv\Scripts\activate  # Windows
# or
source venv/bin/activate  # Mac/Linux

Run tests:

pytest

All tests should pass if your environment is correctly set up.

Tip: If you face an ImportError: No module named 'app', ensure you're in the project root and run:

$env:PYTHONPATH = "."  # Windows PowerShell
# or
export PYTHONPATH=.     # Mac/Linux or bash

📂 Project Structure

├── app/
│   ├── __init__.py          # Flask backend
│   ├── grade_formula.py     # Manual grade calculation
│   └── predictor.py         # Handles model loading and prediction
├── models/                  # Trained .pkl model files
├── static/
│   └── index.js             # Frontend JS logic
├── templates/
│   └── index.html           # Frontend HTML
├── generate_synthetic_data.py
├── train_model.py
├── requirements.txt
└── README.md

Like this project

Posted Jul 6, 2025

Developed a student grading predictor app with secure login and model accuracy visualization.

Likes

Views

Timeline

May 20, 2025 - Jun 18, 2025