FoPra Beluga Challenge - Reinforcement Learning v1.0
Deep Reinforcement Learning solution for the Beluga Challenge shipping container optimization problem using PPO and MCTS
Class List
Here are the classes, structs, unions and interfaces with brief descriptions:
[detail level 12345]
 ▼Nrl
 ▼Nagents
 ▼Nhigh_level
 ▼Nppo_agent
 CActorNetworkActor network for PPO agent
 CCriticNetworkCritic network for PPO agent
 CPPOAgentProximal Policy Optimization (PPO) agent implementation
 CPPOMemoryMemory buffer for storing PPO training experiences
 ▼Nenv
 ▼Nenvironment
 CEnvBeluga Challenge environment for reinforcement learning
 ▼Nstate
 CBelugaContainer ship (Beluga) with current and outgoing jigs
 CJigIndividual jig instance with type and status
 CJigTypeRepresents a type of jig with size properties
 CProblemStateComplete state representation for the Beluga Challenge
 CProductionLineFactory production line with scheduled jigs
 CRackStorage rack with size constraints
 ▼Nmcts
 ▼Nmcts
 CMCTSMonte Carlo Tree Search algorithm implementation
 ▼Nmcts_node
 CMCTSNodeNode class for Monte Carlo Tree Search
 ▼Ntraining
 ▼Ntrainer
 CTrainerMain training orchestrator for the Beluga Challenge