continuous

Source code: tianshou/utils/net/continuous.py

---

class Actor(preprocess_net: Module | Net, action_shape: Sequence[int] | int | int64, hidden_sizes: Sequence[int] = (), max_action: float = 1.0, device: str | int | device = 'cpu', preprocess_net_output_dim: int | None = None)

Simple actor network that directly outputs actions for continuous action space. Used primarily in DDPG and its variants. For probabilistic policies, see ActorProb.

It will create an actor operated in continuous action space with structure of preprocess_net —> action_shape.

Parameters:

• preprocess_net – a self-defined preprocess_net, see usage. Typically, an instance of Net.

• action_shape – a sequence of int for the shape of action.

• hidden_sizes – a sequence of int for constructing the MLP after preprocess_net.

• max_action – the scale for the final action.

• preprocess_net_output_dim – the output dimension of preprocess_net. Only used when preprocess_net does not have the attribute output_dim.

For advanced usage (how to customize the network), please refer to Build the Network.

forward(obs: ndarray | Tensor, state: Any = None, info: dict[str, Any] | None = None) → tuple[Tensor, Any]

Mapping: s_B -> action_values_BA, hidden_state_BH | None.

Returns a tensor representing the actions directly, i.e, of shape (n_actions, ), and a hidden state (which may be None). The hidden state is only not None if a recurrent net is used as part of the learning algorithm (support for RNNs is currently experimental).

get_output_dim() → int

get_preprocess_net() → Module

class ActorProb(preprocess_net: Module | Net, action_shape: Sequence[int] | int | int64, hidden_sizes: Sequence[int] = (), max_action: float = 1.0, device: str | int | device = 'cpu', unbounded: bool = False, conditioned_sigma: bool = False, preprocess_net_output_dim: int | None = None)

Simple actor network that outputs mu and sigma to be used as input for a dist_fn (typically, a Gaussian).

Used primarily in SAC, PPO and variants thereof. For deterministic policies, see Actor.

Parameters:

• preprocess_net – a self-defined preprocess_net, see usage. Typically, an instance of Net.

• action_shape – a sequence of int for the shape of action.

• hidden_sizes – a sequence of int for constructing the MLP after preprocess_net.

• max_action – the scale for the final action logits.

• unbounded – whether to apply tanh activation on final logits.

• conditioned_sigma – True when sigma is calculated from the input, False when sigma is an independent parameter.

• preprocess_net_output_dim – the output dimension of preprocess_net. Only used when preprocess_net does not have the attribute output_dim.

For advanced usage (how to customize the network), please refer to Build the Network.

forward(obs: ndarray | Tensor, state: Any = None, info: dict[str, Any] | None = None) → tuple[tuple[Tensor, Tensor], Any]

Mapping: obs -> logits -> (mu, sigma).

get_output_dim() → int

get_preprocess_net() → Module

class Critic(preprocess_net: ~torch.nn.modules.module.Module | ~tianshou.utils.net.common.Net, hidden_sizes: ~collections.abc.Sequence[int] = (), device: str | int | ~torch.device = 'cpu', preprocess_net_output_dim: int | None = None, linear_layer: ~collections.abc.Callable[[int, int], ~torch.nn.modules.module.Module] = <class 'torch.nn.modules.linear.Linear'>, flatten_input: bool = True)

Simple critic network.

It will create an actor operated in continuous action space with structure of preprocess_net —> 1(q value).

Parameters:

• preprocess_net – a self-defined preprocess_net, see usage. Typically, an instance of Net.

• hidden_sizes – a sequence of int for constructing the MLP after preprocess_net.

• preprocess_net_output_dim – the output dimension of preprocess_net. Only used when preprocess_net does not have the attribute output_dim.

• linear_layer – use this module as linear layer.

• flatten_input – whether to flatten input data for the last layer.

For advanced usage (how to customize the network), please refer to Build the Network.

forward(obs: ndarray | Tensor, act: ndarray | Tensor | None = None, info: dict[str, Any] | None = None) → Tensor

Mapping: (s_B, a_B) -> Q(s, a)_B.

class CriticBase(*args, **kwargs)

abstract forward(obs: ndarray | Tensor, act: ndarray | Tensor | None = None, info: dict[str, Any] | None = None) → Tensor

Mapping: (s_B, a_B) -> Q(s, a)_B.

class Perturbation(preprocess_net: Module, max_action: float, device: str | int | device = 'cpu', phi: float = 0.05)

Implementation of perturbation network in BCQ algorithm.

Given a state and action, it can generate perturbed action.

Parameters:

• preprocess_net – a self-defined preprocess_net which output a flattened hidden state.

• max_action – the maximum value of each dimension of action.

• device – which device to create this model on.

• phi – max perturbation parameter for BCQ.

For advanced usage (how to customize the network), please refer to Build the Network.

See also

You can refer to examples/offline/offline_bcq.py to see how to use it.

forward(state: Tensor, action: Tensor) → Tensor

Defines 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.

class RecurrentActorProb(layer_num: int, state_shape: Sequence[int], action_shape: Sequence[int], hidden_layer_size: int = 128, max_action: float = 1.0, device: str | int | device = 'cpu', unbounded: bool = False, conditioned_sigma: bool = False)

Recurrent version of ActorProb.

For advanced usage (how to customize the network), please refer to Build the Network.

forward(obs: ndarray | Tensor, state: dict[str, Tensor] | None = None, info: dict[str, Any] | None = None) → tuple[tuple[Tensor, Tensor], dict[str, Tensor]]

Almost the same as Recurrent.

class RecurrentCritic(layer_num: int, state_shape: Sequence[int], action_shape: Sequence[int] = [0], device: str | int | device = 'cpu', hidden_layer_size: int = 128)

Recurrent version of Critic.

For advanced usage (how to customize the network), please refer to Build the Network.

forward(obs: ndarray | Tensor, act: ndarray | Tensor | None = None, info: dict[str, Any] | None = None) → Tensor

Almost the same as Recurrent.

class VAE(encoder: Module, decoder: Module, hidden_dim: int, latent_dim: int, max_action: float, device: str | device = 'cpu')

Implementation of VAE.

It models the distribution of action. Given a state, it can generate actions similar to those in batch. It is used in BCQ algorithm.

Parameters:

• encoder – the encoder in VAE. Its input_dim must be state_dim + action_dim, and output_dim must be hidden_dim.

• decoder – the decoder in VAE. Its input_dim must be state_dim + latent_dim, and output_dim must be action_dim.

• hidden_dim – the size of the last linear-layer in encoder.

• latent_dim – the size of latent layer.

• max_action – the maximum value of each dimension of action.

• device – which device to create this model on.

For advanced usage (how to customize the network), please refer to Build the Network.

See also

You can refer to examples/offline/offline_bcq.py to see how to use it.

decode(state: Tensor, latent_z: Tensor | None = None) → Tensor

forward(state: Tensor, action: Tensor) → tuple[Tensor, Tensor, Tensor]

Defines 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.