T3-Neural

T3-Neural is an advanced Tic Tac Toe project that features Reinforcement Learning (RL) bots trained using Q-learning to play as X and O. The project integrates a modern frontend built using Angular 17 with a backend powered by FastAPI for deploying the trained RL models. The RL models are implemented in PyTorch and trained on randomly generated games to master Tic Tac Toe strategies.

Features

• RL-Powered Bots: Intelligent bots for X and O trained using Q-learning.
• Frontend: An interactive and responsive UI built with Angular 17.
• Backend: A FastAPI-based backend for serving the trained PyTorch models.
• Training Pipeline: Models trained on randomly generated games to ensure comprehensive learning of Tic Tac Toe strategies.
• Cross-Platform: Compatible with modern web browsers for a seamless experience.

---

Tech Stack

• Frontend: Angular 17
• Backend: FastAPI
• Model: PyTorch
• Reinforcement Learning: Q-learning algorithm

---

How It Works

1. The frontend allows users to play Tic Tac Toe against RL bots.
2. Bots are served via the backend, which uses PyTorch models trained with Q-learning.
3. The game states are processed in real-time, and bots make moves based on their learned policies.
4. Users can interact with the game and analyze the bot's strategies.

---

Training Details

• The RL bots were trained using the Q-learning algorithm.
• Training involved randomly generated games to explore all possible Tic Tac Toe scenarios.
• PyTorch was used to implement and train the models.

---

Future Enhancements

• Dynamic Difficulty Adjustment: Allow users to set bot difficulty levels.
• Multiplayer Mode: Enable online multiplayer functionality.

---

Contributing

Contributions are welcome! Please follow these steps:

1. Fork the repository.
2. Create a new branch for your feature/fix.
3. Commit your changes and push to your branch.
4. Create a pull request.

---

License

This project is licensed under the MIT License. See the LICENSE file for more details.

---