def reset(self): self.wealth = 10e3 self.total_reward = 0 self.ep_ave_max_q = 0 self.loss = 0 self.actor_loss = 0 self.wealth_history = [] self.r_history = [] self.w_history = [] self.p_history = [] self.noise = OrnsteinUhlenbeckActionNoise(mu=np.zeros(M))
def reset(self): self.wealth = 10e3 self.total_reward = 0 self.ep_ave_max_q = 0 self.loss = 0 self.actor_loss = 0 self.wealth_history = [] self.r_history = [] self.w_history = [] self.p_history = [] self.noise = OrnsteinUhlenbeckActionNoise( mu=np.zeros(M) ) # thêm actionnoise, M ở phía sau sẽ được khai báo bằng số lượng stock