StarClassifier

StarClassifier is a web application that allows users to upload images and classify them using machine learning. It's built with a React frontend, Express backend, and a FastAPI-based machine learning service.

Project Structure

The project is divided into three main parts:

1. Frontend (React)
2. Backend (Express)
3. ML Backend (FastAPI)

(flows folder have the flowchart explanation of the project)

Setup and Installation

1. Clone the repository

2. Install dependencies for each part of the project:

   open 3 terminals
   cd frontend && npm install
   cd backend && npm install
   cd ml-backend && poetry install && poetry shell
   

3. Set up environment variables:

   • Create a .env file in the backend directory with help of .env.example file
   • Create a .env file in the frontend directory with help of .env.example file

4. Start the services:

   • Frontend: cd frontend && npm start
   • Backend: cd backend && npm start
   • ML Backend: cd ml-backend && uvicorn main:app --reload

Contributing

Contributions are welcome! Please see the CONTRIBUTING.md file for guidelines.

Acknowledgements

Developed by the Catalyst community - @leocodeio

DOCKER IMPLEMENTATION

Individual Services

go to /frontend

docker build -t fe-image .
docker run -p 3000:3000 fe-image

you can access the frontend at http://localhost:3000

go to /backend

docker build -t be-image .
docker run -p 3001:3001 be-image

you can access the backend at http://localhost:3001

go to /ml-backend

docker build -t ml-be-image .
docker run -p 8000:8000 ml-be-image

you can access the ml-backend at http://localhost:8000

All Services

in the root directory

docker compose up -d

you can access the frontend at http://localhost:3000 you can access the backend at http://localhost:3001 you can access the ml-backend at http://localhost:8000/docs

to stop the services

docker compose down