Beispiel #1
0
 def getReward(self):
     if self.env.perseus[1] == self.length + 1:
         if abs(self.env.perseus[0] - self.env.goal[0]) == 2:
             # bad choice taken
             self.env.perseus = self.env.goal
             return self.bangPenalty
     return MazeTask.getReward(self)
Beispiel #2
0
 def getReward(self):
     if self.env.perseus[1] == self.length + 1:
         if abs(self.env.perseus[0] - self.env.goal[0]) == 2:
             # bad choice taken
             self.env.perseus = self.env.goal
             return self.bangPenalty
     return MazeTask.getReward(self)
Beispiel #3
0
 def getReward(self):
     if self.bad:
         return self.minReward
     else:
         return MazeTask.getReward(self)
 def getReward(self):
     if self.bad:
         return self.minReward
     else:
         return MazeTask.getReward(self)