default_resource_request

classmethod Trainable.default_resource_request(config: Dict[str, Any]) → PlacementGroupFactory | None

Provides a static resource requirement for the given configuration.

This can be overridden by sub-classes to set the correct trial resource allocation, so the user does not need to.

Examples

>>> from ray.tune import PlacementGroupFactory, Trainable
>>> class MyTrainable(Trainable):
...     @classmethod
...     def default_resource_request(cls, config):
...         return PlacementGroupFactory([{"CPU": 1}, {"CPU": 1}])
>>> MyTrainable.default_resource_request({}).required_resources
{'CPU': 2.0}

Parameters:

config (Dict[str, Any]) – The Trainable’s config dict.

Returns:

A PlacementGroupFactory consumed by Tune

for queueing.

Return type:

PlacementGroupFactory