def after_step()

in robosumo/envs/agents.py [0:0]

• 4 lines of code
• 1 McCabe index (conditional complexity)

    def after_step(self, action):
        self.posafter = self.get_qpos()[:2].copy()
        # Control cost
        reward = - self.COST_COEFS['ctrl'] * np.square(action).sum()
        return reward