AIPokerBot

AIPokerBot is a small Python poker game project with a built-in AI player trained using Counterfactual Regret Minimization (CFR). It provides a playable game environment and AI utilities for experimenting with simple poker agents.

Features

• Play a simplified poker game via main.py.
• AI implementation and training code under the AiPlayer/ package (CFR based).
• Modular game code: Deck.py, Board.py, Player.py, and game_logic.py.

Repository Structure

• main.py — entry point to run the game.
• Guienv.py — (GUI environment) currently in the repo root.
• Board.py, Deck.py, Player.py, game_logic.py — core game logic and models.
• AiPlayer/ — AI code, including CFR implementation, training, and helpers.
  • AiPlayer.py, cfr.py, TrainAi.py, policy_visualizer.py, and cfr_policy.json.

Requirements

• Python 3.8 or newer.
• No external packages are strictly required for the core game; optional packages may be used by AI training visualizers.

Installation

1. Create and activate a virtual environment (recommended):

python -m venv .venv
.\.venv\Scripts\activate

2. (Optional) Install packages if you add dependencies:

pip install -r requirements.txt

Running the Game

Run the main script to start a game session:

python main.py

Training the AI

The AiPlayer/ directory contains code for training and visualizing CFR-based policies. See AiPlayer/TrainAi.py and AiPlayer/cfr.py for details.