FL-ChatBot

FL-ChatBot is a chat application built with React and Flask that allows users to send text messages and images. The application utilizes a federated learning approach for image classification, enabling the chatbot to learn and improve based on user interactions.

Project Overview

• Built with: React (client-side), Flask (server-side)
• Features:
  • Send text messages
  • Send images
  • Image classification with federated learning
  • Real-time chat history and bot responses

Installation

1. Clone the repository:

   git clone <repository_url>

2. Navigate to the project directory:

   cd FL-ChatBot

3. Set up the Flask server:

   a. Navigate to the server directory:

   cd server

   b. Create and activate a virtual environment (optional but recommended):

   python -m venv venv
   source venv/bin/activate  # On Windows, use `venv\Scripts\activate`

   c. Install the required Python packages:

   pip install opencv-python
   pip install flask
   pip install flask-cors
   pip install tensorflow
   pip install keras
   pip install scikit-learn
   pip install flwr
   pip install numpy
   pip install pydicom
   pip install pillow

4. Set up the React client:

   a. Navigate to the client directory:

   cd ../client

   b. Install the required npm packages:

   npm install
   npm install concurrently
   npm install cross-env

Running the Application

To run both the Flask client and the React client simultaneously, use the following command in the client directory:

npm start

To run the federated learning server, use the following command in the server directory:

python server.py

server.py will perform federated learning after minimum 4 client registiration after first upload.

Usage

• Interact with the chat interface to send text messages and images.
• View the chat history and bot responses in real-time.

URLs

• Client URL: http://localhost:3000
• Server URL: http://localhost:5000

Note: The server URL might be used for internal functionalities and might not be relevant to users.