Struct MultiAgentEnvLoop 

pub struct MultiAgentEnvLoop<BT, RLC>
where
    BT: Backend,
    RLC: RLComponentsTypes,
{ /* private fields */ }

An asynchronous runner for multiple agent/environement interfaces.

Implementations

impl<BT, RLC> MultiAgentEnvLoop<BT, RLC>

where BT: Backend, RLC: RLComponentsTypes,

pub fn new( num_envs: usize, env_init: <RLC as RLComponentsTypes>::EnvInit, agent: AsyncPolicy<<RLC as RLComponentsTypes>::Backend, <RLC as RLComponentsTypes>::Policy>, eval: bool, deterministic: bool, device: &<BT as BackendTypes>::Device, ) -> MultiAgentEnvLoop<BT, RLC>

Create a new asynchronous runner for multiple agent/environement interfaces.

Trait Implementations

impl<BT, RLC> AgentEnvLoop<BT, RLC> for MultiAgentEnvLoop<BT, RLC>

where BT: Backend, RLC: RLComponentsTypes,

fn run_steps( &mut self, num_steps: usize, processor: &mut AsyncProcessorTraining<RLEvent<<RLC as RLComponentsTypes>::TrainingOutput, <RLC as RLComponentsTypes>::ActionContext>, AgentEvaluationEvent<<RLC as RLComponentsTypes>::ActionContext>>, interrupter: &Interrupter, progress: &mut Progress, ) -> Vec<TimeStep<BT, <RLC as RLComponentsTypes>::State, <RLC as RLComponentsTypes>::Action, <RLC as RLComponentsTypes>::ActionContext>>

Run a certain number of timesteps.

fn update_policy(&mut self, update: <RLC as RLComponentsTypes>::PolicyState)

Update the runner’s agent.

fn run_episodes( &mut self, num_episodes: usize, processor: &mut AsyncProcessorTraining<RLEvent<<RLC as RLComponentsTypes>::TrainingOutput, <RLC as RLComponentsTypes>::ActionContext>, AgentEvaluationEvent<<RLC as RLComponentsTypes>::ActionContext>>, interrupter: &Interrupter, _progress: &mut Progress, ) -> Vec<Trajectory<BT, <RLC as RLComponentsTypes>::State, <RLC as RLComponentsTypes>::Action, <RLC as RLComponentsTypes>::ActionContext>>

Run a certain number of episodes.

fn policy(&self) -> <RLC as RLComponentsTypes>::PolicyState

Get the state of the runner’s agent.