csle_agents.agents.dqn package

Submodules

csle_agents.agents.dqn.dqn_agent module

class csle_agents.agents.dqn.dqn_agent.DQNAgent(simulation_env_config: SimulationEnvConfig, emulation_env_config: Optional[EmulationEnvConfig], experiment_config: ExperimentConfig, training_job: Optional[TrainingJobConfig] = None, save_to_metastore: bool = True)

Bases: BaseAgent

A DQN agent using the implementation from OpenAI baselines

hparam_names() → List[str]

Gets the hyperparameters

Returns

a list with the hyperparameter names

train() → ExperimentExecution

Implements the training logic of the DQN algorithm

Returns

the experiment result

class csle_agents.agents.dqn.dqn_agent.DQNTrainingCallback(exp_result: ExperimentResult, seed: int, random_seeds: List[int], training_job: TrainingJobConfig, exp_execution: ExperimentExecution, max_steps: int, simulation_name: str, states: List[State], actions: List[Action], player_type: PlayerType, env: BaseEnv, experiment_config: ExperimentConfig, verbose=0, eval_every: int = 100, eval_batch_size: int = 10, save_every: int = 10, save_dir: str = '', L: int = 3, gym_env_name: str = '')

Bases: BaseCallback

Callback for monitoring DQN training

Module contents