How kaleblee146/cecs-491b-project works
Project type and purpose
• This is a full-stack, AI-powered fitness application composed of:
• A Django REST backend (with analytics, user mgmt, notifications, workout planning, content)
• An iOS SwiftUI front-end (MoveMentorApp) including a real-time camera module for pose analysis
• Infrastructure for containerization (Docker, Docker-Compose) and AWS Elastic Beanstalk deployment
File structure
• Top-level directories:
• DjangoBackend/ – API server
– backend/ – Django project bootstrap (settings, URL routing, ASGI/WSGI, WebSocket routing for dash_app)
– analytics/, core/, notifications/, plans/, resources/, users/, workouts/ – Django apps (modular services)
– .platform/nginx/ + Dockerfile + Procfile + docker-compose.yml – deployment config for Nginx, Gunicorn, Docker
• MoveMentorApp/ – iOS app
– Managers/ – network communications (NetworkManager.swift)
– Views/ – SwiftUI screens + Camera_Components for live pose capture & AI inference model (PoseNet)
• Configuration indicates a reverse proxy (Nginx) → WSGI server (Gunicorn) → Django, plus Dockerized local development
README for additional insights
• REST API endpoints exposed at port 8000 for both mobile app and camera streams
• WebSocket support via Django Channels (dash_app consumers & routing) for live dashboards
• AWS EB used as production host; DB credentials managed via environment variables
Detail & modularity
• Camera & AI inference as a sub-component of the mobile app
• Deployment artifacts (Docker Compose services, EB config, Nginx conf)