def get_mdp_refined(self) -> MDPRefined:
return MDPRefined(self.get_mdp_refined_dict(), gamma=1.)
1: (0.2, 4.8),
2: (0.4, 9.2),
3: (0.4, -8.2)
}
},
3: {
'a': {
3: (1.0, 0.0)
},
'b': {
3: (1.0, 0.0)
}
}
}
gamma_val = 1.0
mdp_ref_obj1 = MDPRefined(mdp_refined_data, gamma_val)
mdp_rep_obj = mdp_ref_obj1.get_mdp_rep_for_rl_tabular()
exploring_start_val = False
softmax_flag = False
episodes_limit = 10000
epsilon_val = 0.1
epsilon_half_life_val = 1000
learning_rate_val = 0.1
max_steps_val = 1000
state_ff = [lambda s: float(s)]
sa_ff = [
lambda x: float(x[0]),
lambda x: 1. if x[1] == 'a' else 0.,
lambda x: 1. if x[1] == 'b' else 0.,
lambda x: 1. if x[1] == 'c' else 0.,
def get_mdp_refined(self, model) -> MDPRefined:
return MDPRefined(self.get_mdp_refined_dict(model), self.epoch_disc_factor)