ray.rllib.env.env_runner_group.EnvRunnerGroup.foreach_worker_with_id#
- EnvRunnerGroup.foreach_worker_with_id(func: Callable[[int, EnvRunner], T], *, local_env_runner: bool = True, healthy_only: bool = True, remote_worker_ids: List[int] = None, timeout_seconds: float | None = None, return_obj_refs: bool = False, mark_healthy: bool = False, local_worker=-1) List[T] [source]#
Calls the given function with each EnvRunner and its ID as its arguments.
- Parameters:
func – The function to call for each worker (as only arg).
local_env_runner – Whether to apply
func
tn local worker, too. Default is True.healthy_only – Apply
func
on known-to-be healthy workers only.remote_worker_ids – Apply
func
on a selected set of remote workers.timeout_seconds – Time to wait for results. Default is None.
return_obj_refs – whether to return ObjectRef instead of actual results. Note, for fault tolerance reasons, these returned ObjectRefs should never be resolved with ray.get() outside of this WorkerSet.
mark_healthy – Whether to mark all those workers healthy again that are currently marked unhealthy AND that returned results from the remote call (within the given
timeout_seconds
). Note that workers are NOT set unhealthy, if they simply time out (only if they return a RayActorError). Also note that this setting is ignored ifhealthy_only=True
(b/cmark_healthy
only affects workers that are currently tagged as unhealthy).
- Returns:
The list of return values of all calls to
func([worker, id])
.
DeveloperAPI: This API may change across minor Ray releases.