def get_fa_expected_sarsa_lambda(self) -> tdl_fa.TDLambda:
return tdl_fa.TDLambda(
self.get_mdp_rep_for_rl_fa(),
TDAlgorithm.ExpectedSARSA,
self.softmax,
self.epsilon,
self.epsilon_half_life,
self.learning_rate,
self.lambd,
self.num_episodes,
self.max_steps,
self.fa_spec,
self.tdl_fa_offline
)
def get_fa_qlearning_lambda(self) -> tdl_fa.TDLambda:
return tdl_fa.TDLambda(
self.get_mdp_rep_for_rl_fa(),
self.exploring_start,
TDAlgorithm.QLearning,
self.softmax,
self.epsilon,
self.epsilon_half_life,
self.lambd,
self.num_episodes,
self.batch_size,
self.max_steps,
self.fa_spec,
self.tdl_fa_offline
)