def step(self, actions: List[ActionInt]): for agent in self.agents: agent.starting_actual_phase = agent.actual_phase self.A.append(start_A()) self._pass_actions_to_agents(actions) self._modify_A() Globals().time += 1 self._execute_phase() self.save_motions() rewards = self.count_rewards() for agent in self.agents: agent.remember(densities=self.x[self.t], reward=rewards[agent.index]) self.global_rewards.append(rewards) self._count_cars_out() self.remember_memory()
def step(self, actions: List[ActionInt]): for agent in self.agents: agent.starting_actual_phase = agent.actual_phase self.A.append(start_A()) # self.x[0] = [i for i in range(40)] # w envie dodac! self._pass_actions_to_agents(actions) self._modify_A() # self.update_global_memory_lights() Globals().time += 1 self._execute_phase() # self.x[self.t] = [i for i in range(40)] # w envie dodac! self.save_motions() # rewards = self.count_rewards_negative() rewards = self.count_rewards() for agent in self.agents: agent.remember(densities=self.x[self.t], reward=rewards[agent.index]) self.global_rewards.append(rewards) self._count_cars_out() self.remember_memory()