SharedStateCriticComposer

class maze.perception.models.critics.shared_state_critic_composer.SharedStateCriticComposer(observation_spaces_dict: Dict[Union[str, int], gym.spaces.Dict], agent_counts_dict: Dict[Union[str, int], int], networks: Union[None, Mapping[str, Any], Any], stack_observations: bool)

One critic is shared across all sub-steps or actors (default to use for standard gym-style environments).

Instantiates a TorchSharedStateCritic.

Parameters

• observation_spaces_dict – Dict of sub-step id to observation space.

• networks – The single, shared critic network as defined in the config.

property critic

(overrides BaseStateCriticComposer)

implementation of BaseStateCriticComposer