OakEmu-RL v0.1.0

Appearance

TrajectoryStartDecision ​

python
TrajectoryStartDecision(source_kind: TrajectorySourceKind, recording: TrajectoryRecording | None = None, source_reference: TrajectorySliceReference | None = None, behaviour_kind: TrajectoryBehaviourKind = 'policy', action_budget: int | None = None, initial_frames: bytes | None = None) -> None

Decision returned when choosing how to start a new episode.

Source

Values ​

action_budget ​

python
action_budget: int | None = None

Actions the episode may take before it is abandoned as a loss; None leaves it uncapped.

Source

behaviour_kind ​

python
behaviour_kind: TrajectoryBehaviourKind = 'policy'

Source

initial_frames ​

python
initial_frames: bytes | None = None

Observation frame stack to seed the restart with; None repeats the restored frame.

Source

recording ​

python
recording: TrajectoryRecording | None = None

Source

source_kind ​

python
source_kind: TrajectorySourceKind

Source

source_reference ​

python
source_reference: TrajectorySliceReference | None = None

Source