`evorl.algorithms.contrib.ars_linear`¶

Module Contents¶

Classes¶

Functions¶

`make_deterministic_ec_agent`
`make_policy_network`	Creates a policy network.

API¶

class evorl.algorithms.contrib.ars_linear.ARSWorkflow(*, env: evorl.envs.Env, agent: evorl.agent.Agent, ec_optimizer: evorl.ec.optimizers.EvoOptimizer, ec_evaluator: evorl.evaluators.Evaluator, evaluator: evorl.evaluators.Evaluator, agent_state_vmap_axes: evorl.agent.AgentStateAxis = 0, config: omegaconf.DictConfig)[source]¶

Bases: evorl.algorithms.ec.so.es_workflow.ESWorkflowTemplate

classmethod name()[source]¶

evorl.algorithms.contrib.ars_linear.make_deterministic_ec_agent(action_space: evorl.envs.Space, actor_hidden_layer_sizes: tuple[int] = (256, 256), use_bias: bool = True, norm_layer_type: str = 'none', normalize_obs: bool = False)[source]¶

evorl.algorithms.contrib.ars_linear.make_policy_network(action_size: int, hidden_layer_sizes: collections.abc.Sequence[int] = (256, 256), use_bias: bool = True, activation: evorl.networks.ActivationFn = nn.relu, activation_final: evorl.networks.ActivationFn | None = None, norm_layer_type: str = 'none') → flax.linen.Module[source]¶: Creates a policy network.