Age predictor web application.
Take a picture and see what age you look like.
Neurage is a React TypeScript project, styled using Styled Components. It is organized as a monorepo using Yarn workspaces, with the Python backend containerized using Docker. API requests are handled using FastAPI and the convolutional neural network created using TensorFlow. Other dependencies include Numpy, Pandas, sklearn, and Pydantic.
The dataset and model used for this project can be found here.
Though this project uses Yarn workspaces as its development environment, it uses separate deployment of the frontend and backend; the frontend using Netlify, and the backend in a server, needing only the appropriate value for the environment variable VITE_BACKEND_SERVER for the former and docker compose up to start up the latter.
-
Frontend:
cd packages/client/ && yarn -
Backend:
python3 -m venv venv && source /venv/bin/activate
pip install -r requirements.txt
OR
cd packages/server/app/ && docker compose up
Set the environment variable VITE_BACKEND_URL in packages/client/.env to the appropriate URL.
yarn start - Runs both client and server
yarn client - Runs only the client
yarn server - Runs only the server