AI Game Hub

A collection of classic games played against Reinforcement Learning agents.

Games

1. Tic-Tac-Toe

• Agent: Q-Learning (Table-based).
• Description: The classic 3x3 game. The agent learns from its mistakes and becomes unbeatable.

2. Connect 4

• Agent: Deep Q-Network (DQN) with PyTorch.
• Description: A strategy game on a 7x6 grid. The agent uses a Convolutional Neural Network to evaluate board states.

Setup

1. Clone the repository:

   git clone https://github.com/Venuenugula/Game_Hub.git
   cd Game_Hub

2. Install dependencies:

   pip install -r requirements.txt

3. Run the application:

   python app.py

4. Open your browser and go to http://127.0.0.1:5000.

Deployment

This project is ready for deployment on Render.

• Procfile and render.yaml are included.
• Ensure you select the Python environment.

Project Structure

• app.py: Main Flask application (Game Hub).
• agent.py: Tic-Tac-Toe Q-Learning agent.
• connect4_agent.py: Connect 4 DQN agent.
• connect4_game.py: Connect 4 game logic.
• game_utils.py: Shared utilities.
• templates/: HTML files for the Hub and games.