Update main_train.py

main_train.py

@@ -219,15 +219,6 @@
  # judge if pause the whole system
  if not system.render_obj.pause:
  # 2 choose action acoording to current state
- """
- epsilon = 0.5
- epsilon = epsilon + epsilon * (episode_cnt*100 + step_cnt)/(step_num*episode_num)
- ran = np.random.uniform()
- if ran < epsilon:
- action = agent.choose_action(observersion, greedy=0)
- else:
- action = list(np.random.uniform(-1,1,n_actions))
- """
  action_1 = agent_1.choose_action(observersion_1, greedy=agent_1_param_dic["action_noise_factor"] * math.pow((1-episode_cnt / episode_num), 2))
  action_2 = agent_2.choose_action(observersion_2, greedy=agent_2_param_dic["action_noise_factor"]* math.pow((1-episode_cnt / episode_num), 2))
  if if_BS: