Possible bug:It may be a version problem but i cant fix it

[yanglin-code]

Hi,the first point cause indexerror and it may be just a version problem with numpy or it is possible that there is some error

,and the second point , does s mean state and a mean action？

1. irl-imitation/mdp/value_iteration.py

   Line 486 in bc9e9cd

   expected_value = rewards_expanded[start:end, :] + gamma * values_tmp

2. irl-imitation/mdp/value_iteration.py

   Line 333 in bc9e9cd

   return sum([P_s1_s_a * (self.mdp.get_reward_sas(s, a, s1) + self.gamma * self.values[s1])

[Charlesyyun]

How do you fix the first error?