API Reference

This section provides detailed API documentation for all modules.

• Game Module
  • UnoGame Class
  • Example Usage
• Cards Module
  • Card Class
  • Deck Class
  • Card Constants
  • Example Usage
• Environment Module
  • UnoEnv Class
  • Observation Space
  • Action Space
  • Reward Structure
  • MultiplayerUnoEnv Class
  • Example Usage
• Agents Module
  • Base Agent Class
  • RandomAgent Class
  • SB3Agent Class
  • DQNAgent Class
  • Recurrent Agents
  • Example Usage
  • Agent Comparison Table

Quick Links

Core Modules

• Game Module - UNO game engine

• Cards Module - Card and deck classes

• Environment Module - Gymnasium RL environment

• Agents Module - Agent implementations

Usage Example

from src.game import UnoGame
from src.state_action_reward import UnoEnv
from src.agents import SB3Agent

# Create environment
env = UnoEnv()

# Load agent
agent = SB3Agent("models/selfplay_champion.zip")

# Play a game
obs = env.reset()
done = False

while not done:
    action = agent.select_action(obs, env.get_valid_actions())
    obs, reward, done, truncated, info = env.step(action)

print(f"Game result: {'Win' if reward > 0 else 'Loss'}")