FoPra Beluga Challenge - Reinforcement Learning v1.0
Deep Reinforcement Learning solution for the Beluga Challenge shipping container optimization problem using PPO and MCTS
Class Hierarchy
This inheritance list is sorted roughly, but not completely, alphabetically:
[detail level 12]
 Crl.env.state.BelugaContainer ship (Beluga) with current and outgoing jigs
 Crl.env.environment.EnvBeluga Challenge environment for reinforcement learning
 Crl.env.state.JigIndividual jig instance with type and status
 Crl.env.state.JigTypeRepresents a type of jig with size properties
 Crl.mcts.mcts.MCTSMonte Carlo Tree Search algorithm implementation
 Crl.mcts.mcts_node.MCTSNodeNode class for Monte Carlo Tree Search
 ▼Cnn.Module
 Crl.agents.high_level.ppo_agent.ActorNetworkActor network for PPO agent
 Crl.agents.high_level.ppo_agent.CriticNetworkCritic network for PPO agent
 Crl.agents.high_level.ppo_agent.PPOAgentProximal Policy Optimization (PPO) agent implementation
 Crl.agents.high_level.ppo_agent.PPOMemoryMemory buffer for storing PPO training experiences
 Crl.env.state.ProblemStateComplete state representation for the Beluga Challenge
 Crl.env.state.ProductionLineFactory production line with scheduled jigs
 Crl.env.state.RackStorage rack with size constraints
 Crl.training.trainer.TrainerMain training orchestrator for the Beluga Challenge