def get_tabular_expected_sarsa(self) -> td0_tabular.TD0:
return td0_tabular.TD0(self.get_mdp_rep_for_rl_tabular(),
self.exploring_start, TDAlgorithm.ExpectedSARSA,
self.softmax, self.epsilon,
self.epsilon_half_life, self.learning_rate,
self.learning_rate_decay, self.num_episodes,
self.max_steps)
def get_tabular_qlearning(self) -> td0_tabular.TD0:
return td0_tabular.TD0(
self.get_mdp_rep_for_rl_tabular(),
TDAlgorithm.QLearning,
self.softmax,
self.epsilon,
self.epsilon_half_life,
self.learning_rate,
self.learning_rate_decay,
self.num_episodes,
self.max_steps
)