Shortcuts

DdpgMlpQNet

class torchrl.modules.DdpgMlpQNet(mlp_net_kwargs_net1: dict | None = None, mlp_net_kwargs_net2: dict | None = None, device: DEVICE_TYPING | None = None)[source]

DDPG Q-value MLP class.

Presented in “CONTINUOUS CONTROL WITH DEEP REINFORCEMENT LEARNING”, https://arxiv.org/pdf/1509.02971.pdf

The DDPG Q-value network takes as input an observation and an action, and returns a scalar from it. Because actions are integrated later than observations, two networks are created.

Parameters:
  • mlp_net_kwargs_net1 (dict, optional) –

    kwargs for MLP. Defaults to

    >>> {
    ...     'in_features': None,
    ...     'out_features': 400,
    ...     'depth': 0,
    ...     'num_cells': [],
    ...     'activation_class': nn.ELU,
    ...     'bias_last_layer': True,
    ...     'activate_last_layer': True,
    ...     }
    

  • mlp_net_kwargs_net2

    Defaults to

    >>> {
    ...     'in_features': None,
    ...     'out_features': 1,
    ...     'depth': 1,
    ...     'num_cells': [300, ],
    ...     'activation_class': nn.ELU,
    ...     'bias_last_layer': True,
    ... }
    

  • device (torch.device, optional) – device to create the module on.

Examples

>>> import torch
>>> from torchrl.modules import DdpgMlpQNet
>>> net = DdpgMlpQNet()
>>> print(net)
DdpgMlpQNet(
  (mlp1): MLP(
    (0): LazyLinear(in_features=0, out_features=400, bias=True)
    (1): ELU(alpha=1.0)
  )
  (mlp2): MLP(
    (0): LazyLinear(in_features=0, out_features=300, bias=True)
    (1): ELU(alpha=1.0)
    (2): Linear(in_features=300, out_features=1, bias=True)
  )
)
>>> obs = torch.zeros(1, 32)
>>> action = torch.zeros(1, 4)
>>> value = net(obs, action)
>>> print(value.shape)
torch.Size([1, 1])
forward(observation: Tensor, action: Tensor) Tensor[source]

Define the computation performed at every call.

Should be overridden by all subclasses.

Note

Although the recipe for forward pass needs to be defined within this function, one should call the Module instance afterwards instead of this since the former takes care of running the registered hooks while the latter silently ignores them.

Docs

Access comprehensive developer documentation for PyTorch

View Docs

Tutorials

Get in-depth tutorials for beginners and advanced developers

View Tutorials

Resources

Find development resources and get your questions answered

View Resources