RolloutBufferTrainableDiagnostics
python
RolloutBufferTrainableDiagnostics(values: np.ndarray, returns: np.ndarray, advantages: np.ndarray) -> NoneFlattened trainable rollout statistics used for learner-side diagnostics.
Values
advantages
python
advantages: np.ndarrayFlattened policy advantages for trainable rollout items.
returns
python
returns: np.ndarrayFlattened returns for trainable rollout items.
values
python
values: np.ndarrayFlattened value predictions for trainable rollout items.