ray.serve.schema.RayActorOptionsSchema#

pydantic model ray.serve.schema.RayActorOptionsSchema[source]#

Options with which to start a replica actor.

field accelerator_type: str | None = None#

Forces replicas to run on nodes with the specified accelerator type.See accelerator types.

field fallback_strategy: List[Dict[str, Any]] | None = None#

If specified, expresses soft constraints through a list of decorator options to fall back on when scheduling on a node.

field label_selector: Dict[str, str] | None = None#

If specified, requires that the actor run on a node with the specified labels.

field memory: float | None = None#

Restrict the heap memory usage of each replica. Uses a default if null.

Constraints:

  • ge = 0

field num_cpus: float | None = None#

The number of CPUs required by the deployment’s application per replica. This is the same as a ray actor’s num_cpus. Uses a default if null.

Constraints:

  • ge = 0

field num_gpus: float | None = None#

The number of GPUs required by the deployment’s application per replica. This is the same as a ray actor’s num_gpus. Uses a default if null.

Constraints:

  • ge = 0

field resources: Dict = {}#

The custom resources required by each replica.

field runtime_env: dict = {}#

This deployment’s runtime_env. working_dir and py_modules may contain only remote URIs.

validator runtime_env_contains_remote_uris  »  runtime_env[source]#