QLtrain.py

    def QLtrain(self):
        cum_reward = np.zeros((self.num_episodes))
        for ep in range(self.num_episodes):
             current_state = self.discretize_state(self.env.reset())
            done = False
            while not done:
                #choosing action according to our exploration-exploitation policy
                action = self.choose_action(current_state)
                obs, reward, done, _ = self.env.step(action)
                cum_reward[ep]+=reward
                new_state = self.discretize_state(obs)
                self.QLupdate(current_state, action, reward, new_state)
                current_state = new_state
            self.getEpsilon()
            self.getLR()
        return cum_reward
        print('QL based training is finished!')