Backdoor AI - CoreML Chat Application

A web application that uses Apple's CoreML model to provide AI chat functionality.

Overview

This application consists of:

• Backend: A FastAPI server that loads and uses a CoreML model for processing chat messages
• Frontend: A React application that provides a user-friendly chat interface
• Remote Model Server (New): A separate server that hosts the CoreML model and provides prediction endpoints

Prerequisites

• Python 3.8+
• Node.js 14+
• Docker and Docker Compose (optional, for containerized deployment)

Setup Instructions

1. Clone the Repository

git clone <repository-url>
cd <repository-directory>

2. Set Up the CoreML Model

The application requires a CoreML model file (BERTSQUADFP16.mlmodel) to function. You have two options:

Option 1: Using the Local Model Approach

You need to provide a Dropbox direct download link to your model file.

cd backend
python setup_model.py --dropbox-link "https://www.dropbox.com/your-direct-download-link?dl=1"

This script will:

• Update the Dropbox link in the configuration
• Download the model
• Verify that the model works correctly

Option 2: Using the Remote Model Server Approach (Recommended)

This approach offloads model inference to a separate server, avoiding the need to download the model to the main application server.

1. Deploy the remote model server (see model_server/DEPLOYMENT.md for instructions)
2. Configure the main application to use the remote model server:

   export USE_REMOTE_MODEL_SERVER=true
   export REMOTE_MODEL_SERVER_URL="https://your-model-server-url.com"
   export REMOTE_MODEL_SERVER_API_KEY="your-api-key"

For more details on the remote model server approach, see REMOTE_MODEL_SERVER.md.

3. Start the Backend Server

cd backend
pip install -r requirements.txt
python run.py

The backend server will start at http://localhost:8000

4. Start the Frontend Application

cd frontend
npm install
npm start

The frontend application will start at http://localhost:3000

Docker Deployment

You can also deploy the application using Docker Compose:

docker-compose up -d

This will start both the backend and frontend services in containers.

Troubleshooting

Model Not Loading

If the model is not loading correctly:

1. Verify that the model file exists:

   cd backend
   python verify_model.py

2. Check the backend logs for any errors:

   cd backend
   python run.py

3. Make sure the Dropbox link is a direct download link (ends with ?dl=1)

Chat Not Working

If the chat functionality is not working:

1. Check the browser console for any errors
2. Verify that the backend is running and accessible
3. Check the backend health endpoint: http://localhost:8000/
4. Try restarting both the backend and frontend services

API Documentation

The backend API documentation is available at http://localhost:8000/docs when the server is running.

License

[Your License Information]