PredictionBuilder objects are objects that can be passed to environments designed for customizability. The PredictionBuilder-derived custom classes implement 2 functions, reset() and get([handle]). If predictions are not required in every step or not for all agents, then
- reset() is called after each environment reset, to allow for pre-computing relevant data.
- get() is called whenever an step has to be computed, potentially for each agent independently in case of multi-agent environments.
PredictionBuilder(max_depth: int = 20)¶
PredictionBuilder base class.
Called whenever get_many in the observation build is called.
- handle : int, optional
Handle of the agent for which to compute the observation vector.
A prediction structure, specific to the corresponding environment.
Called after each environment reset.