policy_wrapper#
Source code: tianshou/highlevel/params/policy_wrapper.py
- class PolicyWrapperFactory[source]#
Bases:
Generic[TPolicyOut],ToStringMixin,ABC- abstract create_wrapped_policy(policy: BasePolicy, envs: Environments, optim_factory: OptimizerFactory, device: str | device) TPolicyOut[source]#
- class PolicyWrapperFactoryIntrinsicCuriosity(*, feature_net_factory: IntermediateModuleFactory, hidden_sizes: Sequence[int], lr: float, lr_scale: float, reward_scale: float, forward_loss_weight: float)[source]#
Bases:
PolicyWrapperFactory[ICMPolicy]- create_wrapped_policy(policy: BasePolicy, envs: Environments, optim_factory: OptimizerFactory, device: str | device) ICMPolicy[source]#