ray.rllib.utils.schedules.scheduler.Scheduler#

class ray.rllib.utils.schedules.scheduler.Scheduler(fixed_value_or_schedule: float | List[List[int | float]] | List[Tuple[int, int | float]], *, framework: str = 'torch', device: str | None = None)[source]#

Class to manage a scheduled (framework-dependent) tensor variable.

Uses the PiecewiseSchedule (for maximum configuration flexibility)

DeveloperAPI: This API may change across minor Ray releases.

Methods

__init__

Initializes a Scheduler instance.

get_current_value

Returns the current value (as a tensor variable).

update

Updates the underlying (framework specific) tensor variable.

validate

Performs checking of a certain schedule configuration.