`evorl.algorithms.erl.erl_td3.erl_ga`¶

Module Contents¶

Classes¶

`ERLGAWorkflow`	ERL w/ GA.
`EvaluateMetric`

API¶

class evorl.algorithms.erl.erl_td3.erl_ga.ERLGAWorkflow(**kwargs)[source]¶

Bases: evorl.algorithms.erl.erl_td3.erl_td3_workflow.ERLTD3WorkflowTemplate

ERL w/ GA.

Config:

EC: n actors
RL: k actors + k critics
Shared replay buffer

evaluate(state: evorl.types.State) → tuple[evorl.metrics.MetricBase, evorl.types.State][source]¶

learn(state: evorl.types.State) → evorl.types.State[source]¶

classmethod name()[source]¶

step(state: evorl.types.State) → tuple[evorl.metrics.MetricBase, evorl.types.State][source]¶: The basic step function for the workflow to update agent.

class evorl.algorithms.erl.erl_td3.erl_ga.EvaluateMetric[source]¶

Bases: evorl.metrics.MetricBase

rl_episode_lengths: chex.Array¶: None

rl_episode_returns: chex.Array¶: None