Enforce max_depth in strategy
This commit is contained in:
parent
fa5b4c6c1a
commit
ee230f4be9
|
@ -77,8 +77,8 @@ class LookaheadStrategy:
|
|||
reached, assuming all players are using this Strategy.
|
||||
"""
|
||||
reward = self.game.get_reward(state)
|
||||
if not self.game.is_over(state):
|
||||
action = self.choose_action(state, depth=depth)
|
||||
if (self.max_depth is None or depth <= self.max_depth) and not self.game.is_over(state):
|
||||
action = self.choose_action(state, depth=depth+1)
|
||||
future_state = self.game.get_next_state(state, action)
|
||||
reward += self.get_current_and_future_reward(future_state, depth=depth+1)
|
||||
return reward
|
||||
|
|
Loading…
Reference in New Issue