reference/generated/torchrl.trainers.algorithms.configs.modules.DreamerV3SeededPolicyConfig

Run in Google Colab

Colab

Download Notebook

Notebook

View on GitHub

GitHub

torchrl.trainers.algorithms.configs.modules.DreamerV3SeededPolicyConfig

class torchrl.trainers.algorithms.configs.modules.DreamerV3SeededPolicyConfig(_partial_: bool = False, module: Any = '???', seed: int = '???', _target_: str = 'torchrl.modules.DreamerV3SeededPolicy')

Hydra configuration for DreamerV3SeededPolicy.

Examples

>>> from hydra.utils import instantiate
>>> from torchrl.trainers.algorithms.configs import DreamerV3DiscreteActorConfig, DreamerV3SeededPolicyConfig
>>> config = DreamerV3SeededPolicyConfig(
...     module=DreamerV3DiscreteActorConfig(in_features=6, out_features=3), seed=7,
... )
>>> policy = instantiate(config)
>>> policy.get_extra_state()
{'seed': 7, 'counter': 0}