def _assemble_reward(self):
self._rewards = [
sr.HomeReward(negative=True),
sr.EnemyBaseReward(negative=True),
sr.ViewEnemyReward(weight=10),
sr.EnemyBaseArrivedReward(weight=30)
]
def _assemble_reward(self):
self._forward_rewards = [
sr.HomeReward(negative=True),
sr.EnemyBaseReward(negative=True),
sr.ViewEnemyReward(weight=10),
sr.EnemyBaseArrivedReward(weight=30),
sr.MinDistReward(negative=True)
]
self._backward_rewards = [
sr.HomeReward(back=True, negative=True),
sr.EnemyBaseReward(back=True, negative=True),
sr.HomeArrivedReward(weight=30),
sr.MinDistReward(negative=True)
]
self._final_rewards = [sr.RoundTripFinalReward(weight=50)]