Skip to content
Snippets Groups Projects
bobs_friend_TODO_2.py 250 B
Newer Older
  • Learn to ignore specific revisions
  • tuhe's avatar
    tuhe committed
            terminated = True  
            if a == 0:
                s_next = self.s * 1.1
            else:
                if np.random.rand() < 1/4:
                    s_next = 0
                else:
                    s_next = self.s + 12
            reward = s_next - self.s