beb1fb180aec3b062f0b007afe049cf22a1bbb1c,osim-rl/env/osim.py,OsimEnv,compute_reward,#OsimEnv#,32

Before Change

        return min(x, 2*math.pi-x)

    def compute_reward(self):
        obs = self.get_observation()
//        print (obs[0], obs[1])
        up = (2*(math.pi**2) - angular_dist(obs[0],math.pi)**2 - angular_dist(obs[1],0.0)**2)/(2*math.pi**2)
        still = (obs[2]**2 + obs[3]**2) / 400
//        print still
        return up - still

    def is_done(self):
        return False

After Change

        return min(x, 2*math.pi-x)

    def compute_reward(self):
        return 0.0

    def is_done(self):
        return False

Instances