reagent.gym.policies.scorers package

Submodules

reagent.gym.policies.scorers.continuous_scorer module

reagent.gym.policies.scorers.continuous_scorer.sac_scorer(actor_network: reagent.models.base.ModelBase) Union[Callable[[Any, Optional[torch.Tensor]], Any], Callable[[Any], Any]]

reagent.gym.policies.scorers.discrete_scorer module

reagent.gym.policies.scorers.discrete_scorer.apply_possible_actions_mask(scores: torch.Tensor, possible_actions_mask: Optional[torch.Tensor] = None, invalid_score: float = - inf) torch.Tensor
reagent.gym.policies.scorers.discrete_scorer.discrete_dqn_scorer(q_network: reagent.models.base.ModelBase) Union[Callable[[Any, Optional[torch.Tensor]], Any], Callable[[Any], Any]]
reagent.gym.policies.scorers.discrete_scorer.discrete_dqn_serving_scorer(q_network: torch.nn.modules.module.Module) Union[Callable[[Any, Optional[torch.Tensor]], Any], Callable[[Any], Any]]
reagent.gym.policies.scorers.discrete_scorer.get_parametric_input(max_num_actions: int, obs: reagent.core.types.FeatureData)
reagent.gym.policies.scorers.discrete_scorer.parametric_dqn_scorer(max_num_actions: int, q_network: reagent.models.base.ModelBase) Union[Callable[[Any, Optional[torch.Tensor]], Any], Callable[[Any], Any]]
reagent.gym.policies.scorers.discrete_scorer.parametric_dqn_serving_scorer(max_num_actions: int, q_network: torch.nn.modules.module.Module) Union[Callable[[Any, Optional[torch.Tensor]], Any], Callable[[Any], Any]]

reagent.gym.policies.scorers.slate_q_scorer module

reagent.gym.policies.scorers.slate_q_scorer.slate_q_scorer(num_candidates: int, q_network: reagent.models.base.ModelBase) Union[Callable[[Any, Optional[torch.Tensor]], Any], Callable[[Any], Any]]
reagent.gym.policies.scorers.slate_q_scorer.slate_q_serving_scorer(num_candidates: int, q_network: torch.nn.modules.module.Module) Union[Callable[[Any, Optional[torch.Tensor]], Any], Callable[[Any], Any]]

Module contents