lit_sb3.algorithms.dqn.policies

source module lit_sb3.algorithms.dqn.policies

Classes

• QNetwork — Action-Value (Q-Value) network for DQN.

• DQNPolicy — Policy class with Q-Value Net and target net for DQN.

• CnnPolicy — Policy class for DQN when using images as input.

• MultiInputPolicy — Policy class for DQN when using dict observations as input.

source class QNetwork(observation_space: spaces.Space, action_space: spaces.Discrete, features_extractor: BaseFeaturesExtractor, features_dim: int, net_arch: list[int] | None = None, activation_fn: type[nn.Module] = nn.ReLU, normalize_images: bool = True)

Bases : BasePolicy

Action-Value (Q-Value) network for DQN.

Parameters

• observation_space : spaces.Space — Observation space

• action_space : spaces.Discrete — Action space

• net_arch : list[int] | None — The specification of the policy and value networks.

• activation_fn : type[nn.Module] — Activation function

• normalize_images : bool — Whether to normalize images or not, dividing by 255.0 (True by default)

Attributes

• device : torch.device — Infer which device this policy lives on by inspecting its parameters.

• squash_output : bool — (bool) Getter for squash_output.

Methods

• forward — Predict the q-values.

source method QNetwork.forward(obs: PyTorchObs) → torch.Tensor

Predict the q-values.

Parameters

• obs : PyTorchObs — Observation

Returns

• torch.Tensor — The estimated Q-Value for each action.

source class DQNPolicy(observation_space: spaces.Space, action_space: spaces.Discrete, lr_schedule: Schedule, net_arch: list[int] | None = None, activation_fn: type[nn.Module] = nn.ReLU, features_extractor_class: type[BaseFeaturesExtractor] = FlattenExtractor, features_extractor_kwargs: dict[str, Any] | None = None, normalize_images: bool = True, optimizer_class: type[torch.optim.Optimizer] = torch.optim.Adam, optimizer_kwargs: dict[str, Any] | None = None)

Bases : BasePolicy

Policy class with Q-Value Net and target net for DQN.

Parameters

• observation_space : spaces.Space — Observation space

• action_space : spaces.Discrete — Action space

• lr_schedule : Schedule — Learning rate schedule (could be constant)

• net_arch : list[int] | None — The specification of the policy and value networks.

• activation_fn : type[nn.Module] — Activation function

• features_extractor_class : type[BaseFeaturesExtractor] — Features extractor to use.

• features_extractor_kwargs : dict[str, Any] | None — Keyword arguments to pass to the features extractor.

• normalize_images : bool — Whether to normalize images or not, dividing by 255.0 (True by default)

• optimizer_class : type[torch.optim.Optimizer] — The optimizer to use, torch.optim.Adam by default

• optimizer_kwargs : dict[str, Any] | None — Additional keyword arguments, excluding the learning rate, to pass to the optimizer

Attributes

• device : torch.device — Infer which device this policy lives on by inspecting its parameters.

• squash_output : bool — (bool) Getter for squash_output.

Methods

• make_q_net

• forward

• set_training_mode — Put the policy in either training or evaluation mode.

source method DQNPolicy.make_q_net() → QNetwork

source method DQNPolicy.forward(obs: PyTorchObs, deterministic: bool = True) → torch.Tensor

source method DQNPolicy.set_training_mode(mode: bool) → None

Put the policy in either training or evaluation mode.

This affects certain modules, such as batch normalisation and dropout.

Parameters

• mode : bool — if true, set to training mode, else set to evaluation mode

source class CnnPolicy(observation_space: spaces.Space, action_space: spaces.Discrete, lr_schedule: Schedule, net_arch: list[int] | None = None, activation_fn: type[nn.Module] = nn.ReLU, features_extractor_class: type[BaseFeaturesExtractor] = NatureCNN, features_extractor_kwargs: dict[str, Any] | None = None, normalize_images: bool = True, optimizer_class: type[torch.optim.Optimizer] = torch.optim.Adam, optimizer_kwargs: dict[str, Any] | None = None)

Bases : DQNPolicy

Policy class for DQN when using images as input.

Parameters

• observation_space : spaces.Space — Observation space

• action_space : spaces.Discrete — Action space

• lr_schedule : Schedule — Learning rate schedule (could be constant)

• net_arch : list[int] | None — The specification of the policy and value networks.

• activation_fn : type[nn.Module] — Activation function

• features_extractor_class : type[BaseFeaturesExtractor] — Features extractor to use.

• normalize_images : bool — Whether to normalize images or not, dividing by 255.0 (True by default)

• optimizer_class : type[torch.optim.Optimizer] — The optimizer to use, torch.optim.Adam by default

• optimizer_kwargs : dict[str, Any] | None — Additional keyword arguments, excluding the learning rate, to pass to the optimizer

Attributes

• device : torch.device — Infer which device this policy lives on by inspecting its parameters.

• squash_output : bool — (bool) Getter for squash_output.

source class MultiInputPolicy(observation_space: spaces.Dict, action_space: spaces.Discrete, lr_schedule: Schedule, net_arch: list[int] | None = None, activation_fn: type[nn.Module] = nn.ReLU, features_extractor_class: type[BaseFeaturesExtractor] = CombinedExtractor, features_extractor_kwargs: dict[str, Any] | None = None, normalize_images: bool = True, optimizer_class: type[torch.optim.Optimizer] = torch.optim.Adam, optimizer_kwargs: dict[str, Any] | None = None)

Bases : DQNPolicy

Policy class for DQN when using dict observations as input.

Parameters

• observation_space : spaces.Dict — Observation space

• action_space : spaces.Discrete — Action space

• lr_schedule : Schedule — Learning rate schedule (could be constant)

• net_arch : list[int] | None — The specification of the policy and value networks.

• activation_fn : type[nn.Module] — Activation function

• features_extractor_class : type[BaseFeaturesExtractor] — Features extractor to use.

• normalize_images : bool — Whether to normalize images or not, dividing by 255.0 (True by default)

• optimizer_class : type[torch.optim.Optimizer] — The optimizer to use, torch.optim.Adam by default

• optimizer_kwargs : dict[str, Any] | None — Additional keyword arguments, excluding the learning rate, to pass to the optimizer

Attributes

• device : torch.device — Infer which device this policy lives on by inspecting its parameters.

• squash_output : bool — (bool) Getter for squash_output.