# {py:mod}`evorl.algorithms.ddpg`

```{py:module} evorl.algorithms.ddpg
```

```{autodoc2-docstring} evorl.algorithms.ddpg
:parser: autodoc2_docstrings_parser
:allowtitles:
```

## Module Contents

### Classes

````{list-table}
:class: autosummary longtable
:align: left

* - {py:obj}`DDPGAgent <evorl.algorithms.ddpg.DDPGAgent>`
  - ```{autodoc2-docstring} evorl.algorithms.ddpg.DDPGAgent
    :parser: autodoc2_docstrings_parser
    :summary:
    ```
* - {py:obj}`DDPGNetworkParams <evorl.algorithms.ddpg.DDPGNetworkParams>`
  - ```{autodoc2-docstring} evorl.algorithms.ddpg.DDPGNetworkParams
    :parser: autodoc2_docstrings_parser
    :summary:
    ```
* - {py:obj}`DDPGTrainMetric <evorl.algorithms.ddpg.DDPGTrainMetric>`
  -
* - {py:obj}`DDPGWorkflow <evorl.algorithms.ddpg.DDPGWorkflow>`
  -
````

### Functions

````{list-table}
:class: autosummary longtable
:align: left

* - {py:obj}`make_mlp_ddpg_agent <evorl.algorithms.ddpg.make_mlp_ddpg_agent>`
  - ```{autodoc2-docstring} evorl.algorithms.ddpg.make_mlp_ddpg_agent
    :parser: autodoc2_docstrings_parser
    :summary:
    ```
````

### API

`````{py:class} DDPGAgent
:canonical: evorl.algorithms.ddpg.DDPGAgent

Bases: {py:obj}`evorl.agent.Agent`

```{autodoc2-docstring} evorl.algorithms.ddpg.DDPGAgent
:parser: autodoc2_docstrings_parser
```

````{py:method} actor_loss(agent_state: evorl.agent.AgentState, sample_batch: evorl.sample_batch.SampleBatch, key: chex.PRNGKey) -> evorl.types.LossDict
:canonical: evorl.algorithms.ddpg.DDPGAgent.actor_loss

```{autodoc2-docstring} evorl.algorithms.ddpg.DDPGAgent.actor_loss
:parser: autodoc2_docstrings_parser
```

````

````{py:attribute} actor_network
:canonical: evorl.algorithms.ddpg.DDPGAgent.actor_network
:type: flax.linen.Module
:value: >
   None

```{autodoc2-docstring} evorl.algorithms.ddpg.DDPGAgent.actor_network
:parser: autodoc2_docstrings_parser
```

````

````{py:method} compute_actions(agent_state: evorl.agent.AgentState, sample_batch: evorl.sample_batch.SampleBatch, key: chex.PRNGKey) -> tuple[evorl.types.Action, evorl.types.PolicyExtraInfo]
:canonical: evorl.algorithms.ddpg.DDPGAgent.compute_actions

````

````{py:method} critic_loss(agent_state: evorl.agent.AgentState, sample_batch: evorl.sample_batch.SampleBatch, key: chex.PRNGKey) -> evorl.types.LossDict
:canonical: evorl.algorithms.ddpg.DDPGAgent.critic_loss

```{autodoc2-docstring} evorl.algorithms.ddpg.DDPGAgent.critic_loss
:parser: autodoc2_docstrings_parser
```

````

````{py:attribute} critic_network
:canonical: evorl.algorithms.ddpg.DDPGAgent.critic_network
:type: flax.linen.Module
:value: >
   None

```{autodoc2-docstring} evorl.algorithms.ddpg.DDPGAgent.critic_network
:parser: autodoc2_docstrings_parser
```

````

````{py:attribute} discount
:canonical: evorl.algorithms.ddpg.DDPGAgent.discount
:type: float
:value: >
   1

```{autodoc2-docstring} evorl.algorithms.ddpg.DDPGAgent.discount
:parser: autodoc2_docstrings_parser
```

````

````{py:method} evaluate_actions(agent_state: evorl.agent.AgentState, sample_batch: evorl.sample_batch.SampleBatch, key: chex.PRNGKey) -> tuple[evorl.types.Action, evorl.types.PolicyExtraInfo]
:canonical: evorl.algorithms.ddpg.DDPGAgent.evaluate_actions

````

````{py:attribute} exploration_epsilon
:canonical: evorl.algorithms.ddpg.DDPGAgent.exploration_epsilon
:type: float
:value: >
   0.5

```{autodoc2-docstring} evorl.algorithms.ddpg.DDPGAgent.exploration_epsilon
:parser: autodoc2_docstrings_parser
```

````

````{py:method} init(obs_space: evorl.envs.Space, action_space: evorl.envs.Space, key: chex.PRNGKey) -> evorl.agent.AgentState
:canonical: evorl.algorithms.ddpg.DDPGAgent.init

```{autodoc2-docstring} evorl.algorithms.ddpg.DDPGAgent.init
:parser: autodoc2_docstrings_parser
```

````

````{py:property} normalize_obs
:canonical: evorl.algorithms.ddpg.DDPGAgent.normalize_obs

```{autodoc2-docstring} evorl.algorithms.ddpg.DDPGAgent.normalize_obs
:parser: autodoc2_docstrings_parser
```

````

````{py:attribute} obs_preprocessor
:canonical: evorl.algorithms.ddpg.DDPGAgent.obs_preprocessor
:type: typing.Any
:value: >
   'pytree_field(...)'

```{autodoc2-docstring} evorl.algorithms.ddpg.DDPGAgent.obs_preprocessor
:parser: autodoc2_docstrings_parser
```

````

`````

`````{py:class} DDPGNetworkParams
:canonical: evorl.algorithms.ddpg.DDPGNetworkParams

Bases: {py:obj}`evorl.types.PyTreeData`

```{autodoc2-docstring} evorl.algorithms.ddpg.DDPGNetworkParams
:parser: autodoc2_docstrings_parser
```

````{py:attribute} actor_params
:canonical: evorl.algorithms.ddpg.DDPGNetworkParams.actor_params
:type: evorl.types.Params
:value: >
   None

```{autodoc2-docstring} evorl.algorithms.ddpg.DDPGNetworkParams.actor_params
:parser: autodoc2_docstrings_parser
```

````

````{py:attribute} critic_params
:canonical: evorl.algorithms.ddpg.DDPGNetworkParams.critic_params
:type: evorl.types.Params
:value: >
   None

```{autodoc2-docstring} evorl.algorithms.ddpg.DDPGNetworkParams.critic_params
:parser: autodoc2_docstrings_parser
```

````

````{py:attribute} target_actor_params
:canonical: evorl.algorithms.ddpg.DDPGNetworkParams.target_actor_params
:type: evorl.types.Params
:value: >
   None

```{autodoc2-docstring} evorl.algorithms.ddpg.DDPGNetworkParams.target_actor_params
:parser: autodoc2_docstrings_parser
```

````

````{py:attribute} target_critic_params
:canonical: evorl.algorithms.ddpg.DDPGNetworkParams.target_critic_params
:type: evorl.types.Params
:value: >
   None

```{autodoc2-docstring} evorl.algorithms.ddpg.DDPGNetworkParams.target_critic_params
:parser: autodoc2_docstrings_parser
```

````

`````

`````{py:class} DDPGTrainMetric
:canonical: evorl.algorithms.ddpg.DDPGTrainMetric

Bases: {py:obj}`evorl.metrics.MetricBase`

````{py:attribute} actor_loss
:canonical: evorl.algorithms.ddpg.DDPGTrainMetric.actor_loss
:type: chex.Array
:value: >
   None

```{autodoc2-docstring} evorl.algorithms.ddpg.DDPGTrainMetric.actor_loss
:parser: autodoc2_docstrings_parser
```

````

````{py:attribute} critic_loss
:canonical: evorl.algorithms.ddpg.DDPGTrainMetric.critic_loss
:type: chex.Array
:value: >
   None

```{autodoc2-docstring} evorl.algorithms.ddpg.DDPGTrainMetric.critic_loss
:parser: autodoc2_docstrings_parser
```

````

````{py:attribute} raw_loss_dict
:canonical: evorl.algorithms.ddpg.DDPGTrainMetric.raw_loss_dict
:type: evorl.types.LossDict
:value: >
   'metric_field(...)'

```{autodoc2-docstring} evorl.algorithms.ddpg.DDPGTrainMetric.raw_loss_dict
:parser: autodoc2_docstrings_parser
```

````

`````

`````{py:class} DDPGWorkflow(env: evorl.envs.Env, agent: evorl.agent.Agent, optimizer: optax.GradientTransformation, evaluator: evorl.evaluators.Evaluator, replay_buffer: evorl.replay_buffers.AbstractReplayBuffer, config: omegaconf.DictConfig)
:canonical: evorl.algorithms.ddpg.DDPGWorkflow

Bases: {py:obj}`evorl.algorithms.offpolicy_utils.OffPolicyWorkflowTemplate`

````{py:method} name()
:canonical: evorl.algorithms.ddpg.DDPGWorkflow.name
:classmethod:

````

````{py:method} step(state: evorl.types.State) -> tuple[evorl.metrics.MetricBase, evorl.types.State]
:canonical: evorl.algorithms.ddpg.DDPGWorkflow.step

````

`````

````{py:function} make_mlp_ddpg_agent(action_space: evorl.envs.Space, critic_hidden_layer_sizes: tuple[int] = (256, 256), actor_hidden_layer_sizes: tuple[int] = (256, 256), discount: float = 1, exploration_epsilon: float = 0.5, normalize_obs: bool = False, policy_obs_key: str = '', value_obs_key: str = '')
:canonical: evorl.algorithms.ddpg.make_mlp_ddpg_agent

```{autodoc2-docstring} evorl.algorithms.ddpg.make_mlp_ddpg_agent
:parser: autodoc2_docstrings_parser
```
````