reference/generated/torchrl.trainers.algorithms.configs.modules.DreamerV3MLPConfig

Run in Google Colab

Colab

Download Notebook

Notebook

View on GitHub

GitHub

torchrl.trainers.algorithms.configs.modules.DreamerV3MLPConfig

class torchrl.trainers.algorithms.configs.modules.DreamerV3MLPConfig(_partial_: bool = False, in_features: int = '???', out_features: int | None = '???', depth: int = 3, num_cells: int = 1024, outscale: float = 1.0, norm_eps: float = 0.0001, device: Any = None, _target_: str = 'torchrl.modules.DreamerV3MLP')

A class to configure a DreamerV3 multilayer perceptron.

Example

>>> import torch
>>> from hydra.utils import instantiate
>>> from torchrl.trainers.algorithms.configs import DreamerV3MLPConfig
>>> cfg = DreamerV3MLPConfig(
...     in_features=6, out_features=4, depth=2, num_cells=8
... )
>>> net = instantiate(cfg)
>>> y = net(torch.randn(3, 2), torch.randn(3, 4))
>>> assert y.shape == (3, 4)

See also

DreamerV3MLP