Multi-Client AI Chatbot System

This project implements a multi-client chatbot system with a Python server powered by Hugging Face Transformers and a Python client for user interaction. The server handles multiple concurrent clients using multi-threading, and the client provides a command-line interface with mood-based interactions.

Features

• Client-server architecture over TCP
• AI chatbot powered by Hugging Face DialoGPT-medium
• Multi-threaded server for concurrent client support
• Commands: /help, /exit, /clear, /sysinfo
• Mood options: default, sarcastic, enthusiastic, serious
• Monorepo structure

Directory Structure

• server/:
  • src/: Server source code
  • tests/: Server unit tests
  • requirements.txt: Server dependencies
• client_python/:
  • src/: Client source code
  • tests/: Client unit tests
  • requirements.txt: Client dependencies
• shared_docs/: Architecture and protocol documentation
• scripts/: Utility scripts for running components

Prerequisites

• Python 3.8+
• Git
• (Optional) Conda (if using Conda for environment management)

Setup

Using Virtual Environments

1. Clone the Repository:

   git clone https://github.com/moaz-loaie/chatbot.git
   cd chatbot

2. Server Setup:

   • Unix-like systems:

     cd server
     python -m venv venv_server
     source venv_server/bin/activate
     pip install -r requirements.txt
     cd ..

   • Windows:

     cd server
     python -m venv venv_server
     venv_server\Scripts\activate
     pip install -r requirements.txt
     cd ..

3. Client Setup:

   • Unix-like systems:

     cd client_python
     python -m venv venv_client
     source venv_client/bin/activate
     pip install -r requirements.txt
     cd ..

   • Windows:

     cd client_python
     python -m venv venv_client
     venv_client\Scripts\activate
     pip install -r requirements.txt
     cd ..

Using Conda

If you prefer to use Conda for managing environments, follow these steps:

Server Setup with Conda

conda env create -p server_env -f environment.yml
conda activate server_env

Client Setup with Conda

conda env create -p client_env -f environment.yml
conda activate client_env

Note: Ensure you have Conda installed. If not, refer to the Conda installation guide.

Running the Application

Using Virtual ENvironments

1. Start the Server:

   • Unix-like systems:

     cd server
     source venv_server/bin/activate
     python src/main_server.py

   • Windows:

     cd server
     venv_server\Scripts\activate
     python src/main_server.py

2. Start the Client:

   • In a new terminal:

     • Unix-like systems:

       cd client_python
       source venv_client/bin/activate
       python src/main_client.py

     • Windows:

       cd client_python
       venv_client\Scripts\activate
       python src/main_client.py

Using COnda

1. Start the Server:

   • Activate the server environment:

     conda activate server_env

   • Run the server:

     python server/src/main_server.py

2. Start the Client:

   • In a new terminal, activate the client environment:

     conda activate client_env

   • Run the client:

     python client_python/src/main_client.py

Using Scripts (Optional)

Alternatively, you can use the provided scripts for convenience:

• Unix-like systems:

  • Start the server: ./scripts/run_server.sh
  • Start the client: ./scripts/run_python_client.sh

• Windows:

  • Start the server: scripts\run_server.bat
  • Start the client: scripts\run_python_client.bat

Note: Ensure you run the server before starting any clients.

Interaction

• Type messages or commands (e.g., /help).
• Select a mood when prompted.

Running Tests

The project includes unit tests for both the server and client components. These tests use Python's built-in unittest module, so no additional libraries are required beyond the standard Python installation.

Server Tests

1. Navigate to the server/ directory.

2. Activate the server environment:

   • Virtual environment: source venv_server/bin/activate (Unix) or venv_server\Scripts\activate (Windows)
   • Conda: conda activate server_env

3. Run the tests:

   python -m unittest discover tests

Client Tests

1. Navigate to the client_python/ directory.

2. Activate the client environment:

   • Virtual environment: source venv_client/bin/activate (Unix) or venv_client\Scripts\activate (Windows)
   • Conda: conda activate client_env

3. Run the tests:

   python -m unittest discover tests

Notes

• Logs are saved in logs/ directories within server/ and client_python/.
• Ensure the server is running before starting clients.
• The AI model is loaded once at server startup for efficiency, as implemented in ai_model_interface.py.