ray.rllib.env.multi_agent_episode.MultiAgentEpisode.get_agents_to_act

MultiAgentEpisode.get_agents_to_act() → Set[Any]

Returns a set of agent IDs required to send an action to env.step() next.

Those are generally the agents that received an observation in the most recent env.step() call.

Returns:

A set of AgentIDs that are supposed to send actions to the next env.step() call.