Attendance System MVP

Milestone 1 includes FastAPI, SQLAlchemy, SQLite, bcrypt password hashing, and login flows for faculty and students.

MVP v3 adds AI attendance summaries, attendance analytics, low-attendance warnings, PDF report generation, email automation logs, and HoD dashboards.

Setup

cd attendance_system
python -m venv .venv
.venv\Scripts\Activate.ps1
pip install -r requirements.txt

Seed Demo Users

cd backend
python seed.py

Demo credentials:

• Faculty: faculty / faculty123
• HoD: hod / hod123
• Student: 22EC001 / 22EC001

Run Backend

cd backend
python -m uvicorn app.main:app --reload --host 127.0.0.1 --port 8010

API docs: http://127.0.0.1:8010/docs

Run Frontend

cd frontend
streamlit run app.py

Frontend: http://localhost:8501

If you choose a different backend port, start Streamlit with the matching URL:

$env:API_BASE_URL="http://127.0.0.1:8080"
streamlit run app.py

MVP v3 Features

• AI reports: session summaries, low-attendance messages, department summaries.
• Analytics: attendance percentage, 30-day trend comparison, low-attendance detection below 75%.
• Reports: daily/session and weekly department PDFs, stored under reports/YYYY/MM/ in MinIO when MinIO is available.
• Email automation: session closure emails, low-attendance warnings, and weekly HoD report logs. If SMTP is not configured, messages are logged with skipped status for MCP delivery.
• HoD role: login with hod / hod123 after running python seed.py.

Local Hugging Face generation is off by default to keep laptop startup fast. Set AI_ENABLE_LOCAL_MODEL=true and choose HF_MODEL_NAME to enable local text generation.