Filter löschen
Filter löschen

How can I extract the actor-critic model from a saved pre-trained agent?

1 Ansicht (letzte 30 Tage)
I have trained a PPO agent and saved the TrainStat at the end of the training (save TrainStat). I have used the following set of commands to retrieve the actor net but the result I am getting is blank.
TrainStat.agent4
ans =
rlPPOAgent with properties:
AgentOptions: [1×1 rl.option.rlPPOAgentOptions]
>> actor = getActor(TrainStat.agent4)
actor =
rlStochasticActorRepresentation with properties:
ActionInfo: [1×1 rl.util.rlFiniteSetSpec]
ObservationInfo: [1×1 rl.util.rlNumericSpec]
Options: [1×1 rl.option.rlRepresentationOptions]
>> A_net = getModel(actor)
A_net =
[]
Please suggest if I am making any mistakes while saving the TrainStat or is there any other way to proceed.

Antworten (0)

Kategorien

Mehr zu Policies and Value Functions finden Sie in Help Center und File Exchange

Community Treasure Hunt

Find the treasures in MATLAB Central and discover how the community can help you!

Start Hunting!

Translated by